Alex Lascarides

Publications

Books

Bender, E. and A. Lascarides (2019) Linguistic Fundamentals for Natural Language Processing II: 100 Essentials from Semantics and Pragmatics, Morgan and Claypool Publishers.

Available from or
Meaning is a fundamental concept in Natural Language Processing (NLP), in the tasks of both Natural Language Understanding (NLU) and Natural Language Generation (NLG). This is because the aims of these fields are to build systems that understand what people mean when they speak or write, and that can produce linguistic strings that successfully express to people the intended content. In order for NLP to scale beyond partial, task-specific solutions, researchers in these fields must be informed by what is known about how humans use language to express and understand communicative intents. The purpose of this book is to present a selection of useful information about semantics and pragmatics, as understood in linguistics, in a way that's accessible to and useful for NLP practitioners with minimal (or even no) prior training in linguistics.
```
@book{bender:lascarides:2019,
author = {Emily Bender and Alex Lascarides},
year = {2019},
title={Linguistic Fundamentals in Natural Language Processing II: 100 Essentials in Semantics and Pragmatics},
publisher = {Morgan and Claypool Publishers}
}
```
Stenning, K., A. Lascarides and J. Calder (2006) Introduction to Cognition and Communication, MIT Press.

Available from
This introduction to the interdisciplinary study of cognition takes the novel approach of bringing several disciplines to bear on the subject of communication. Using the prespectives of linguistics, logic, AI, philosophy, and psychology---the component fields of cognitive science---to explore topics in human communication in depth, the book shows readers and students from any background how these disciplines developed their distinctive views, and how those views interact. The book introduces some sample phenomena of human communication that illustrate the approach of cognitive science in understanding the mind, and then considers theoretical issues, including the relation of logic and computation and the concept of representation. It describes the development of a model of natural alngauge and explores the link between an utterances and its meaning and how this can be described in a formal way on the basis of recent advances in AI research. It looks at communication employing graphical messages and the similarities and differences between language and diagrams. Finally, the book considers some general philosophical critiques of computational models of mind. The book can be used at a number of different levels. A glossary, suggestions for further reading, and a Web site with multiple-choice questions are provided for nonspecialist students; advanced students can supplement the material with readings that take the topics into greater depth.
```
@book{stenning:etal:2006,
author = {Keith Stenning and Alex Lascarides and Jo Calder},
year = {2006},
title={Introduction to Cognition and Communication},
publisher = {MIT Press}
}
```
Asher, N. and A. Lascarides (2003) Logics of Conversation, Cambridge University Press.

Available from
People often mean more than they say. Grammar on its own is typically insufficient for determining the full meaning of an utterance; the assumption that the discourse is coherent or ‘makes sense’ has an important role to play in determining meaning as well. Logics of Conversation presents a dynamic semantic framework called Segmented Discourse Representation Theory, or SDRT, where this interaction between discourse coherence and discourse interpretation is explored in a logically precise manner. Combining ideas from dynamic semantics, commonsense reasoning and speech act theory, SDRT uses its analysis of rhetorical relations to capture intuitively compelling implicatures. It provides a computable method for constructing these logical forms and is one of the most formally precise and linguistically grounded accounts of discourse interpretation currently available. The book will be of interest to researchers and students in linguistics and in philosophy of language.
```
@book{asher:lascarides:2003,
author = {N. Asher and A. Lascarides},
year = {2003},
title = {Logics of Conversation},
publisher = {Cambridge University Press}
}
```

Journal Articles

Appelgren, M. and A. Lascarides (2020) Interactive Task Learning via Embodied Corrective Feedback, Journal of Autonomous Agents and Multiagent Systems (JAAMAS), doi:https://doi.org/10.1007/s10458-020-09481-8.

This paper addresses a task in Interactive Task Learning (Laird et al. IEEE Intell Syst 32:6–21, 2017). The agent must learn to build towers which are constrained by rules, and whenever the agent performs an action which violates a rule the teacher provides verbal corrective feedback: e.g. “No, red blocks should be on blue blocks”. The agent must learn to build rule compliant towers from these corrections and the context in which they were given. The agent is not only ignorant of the rules at the start of the learning process, but it also has a deficient domain model, which lacks the concepts in which the rules are expressed. Therefore an agent that takes advantage of the linguistic evidence must learn the denotations of neologisms and adapt its conceptualisation of the planning domain to incorporate those denotations. We show that by incorporating constraints on interpretation that are imposed by discourse coherence into the models for learning (Hobbs in On the coherence and structure of discourse, Stanford University, Stanford, 1985; Asher et al. in Logics of conversation, Cambridge University Press, Cambridge, 2003), an agent which utilizes linguistic evidence outperforms a strong baseline which does not.
```
@article{appelgren:lascarides:2020,
author = {M.\ Appelgren and A.\ Lascarides},
year = {2020},
title = {Interactive Task Learning via Corrective Feedback}, 
journal = {Journal of Autonomous Agents and Multi-Agent Systems},
volume = {34},
number = {54},
doi = {doi.org/10.1007/s10458-020-09481-8}
}
```

Schloeder, J. and A. Lascarides (2020) Understanding Focus: Pitch, Placement and Coherence, Semantics and Pragmatics, 1:1--53, doi:http://dx.doi.org/10.3765/sp.13.1.

This paper presents a novel account of focal stress and pitch contour in English dialogue. We argue that one should analyse and treat focus and pitch contour jointly, since (i) some pragmatic interpretations vary with contour (e.g., whether an utterance accepts or rejects; or whether it implicates a positive or negative answer); and (ii) there are utterances with identical prosodic focus that in the same context are infelicitous with one contour, but felicitous with another. We offer an account of two distinct pitch contours that predicts the correct felicity judgements and implicatures, outclassing other models in empirical coverage or formality. Prosodic focus triggers a presupposition, where what is presupposed and how the presupposition is resolved depends on prosodic contour. If resolving the presupposition entails the proffered content, then the proffered content is uninteresting and hence the utterance is infelicitous. Otherwise, resolving the presupposition may lead to an implicature. We regiment this account in SDRT.
```
@article{schloeder:lascarides:2020,
author = {J.\ Schl\"oder and A.\ Lascarides},
year = {2020},
title = {Understanding Focus: Pitch, Placement and Coherence}, 
journal = {Semantics and Pragmatics},
volume = {13},
number = {1},
pages = {1--53},
doi = {dx.doi.org/10.3765/sp.13.1}
}
```
Lascarides, A. and M. Guhe (2019) Persuasion with Limited Sight, Review of Philosophy and Psychology, 10(1), pp1--33, Springer.

Humans face many game problems that are too large for the whole game tree to be used in their deliberations about action, and very little is understood about how they cope in such scenarios. However, when a human player's chosen strategy is conditioned on her limited perspective of how the game might progress (Degremont et al, 2016), then it should be possible to manipulate her into changing her planned move by mentioning a possible outcome of an alternative move. This paper demonstrates that human players can be manipulated this way: in the game The Settlers of Catan, where negotiation is only a small part of what one must do to win the game thereby generating uncertainty about which outcomes to the negotiation are good and which are bad, the likelihood that a player accepts a trade offer that deviates from their declared preferred strategy is higher if it is accompanied by a description of what that trade offer can lead to.
```
@article{lascarides:guhe:2019,
author = {A.\ Lascarides and M.\ Guhe},
year = {2019},
title = {Persuasion with Limited Sight}, 
journal = {Review of Philosophy and Psychology},
volume = {10},
number = {1},
pages = {1--33},
doi = {10.1007/s13164-018-0398-z},
publisher = {Springer}
}
```
Hunter, J. N. Asher, and A. Lascarides (2018) A Formal Semantics for Situated Conversation, Semantics and Pragmatics, doi.org/10.3765/sp.11.10.

While linguists and philosophers have sought to model the various ways in which the meaning of what we say can depend on the nonlinguistic context, this work has by and large focused on how the nonlinguistic context can be exploited to ground or anchor referential or otherwise context-sensitive expressions. In this paper, we focus on examples in which nonlinguistic events contribute entire discourse units that serve as arguments to coherence relations, without the mediation of context-sensitive expressions. We use both naturally occurring and constructed examples to highlight these interactions and to argue that extant coherence-based accounts of discourse should be extended to model them. We also argue that extending coherence-based accounts in this way is a nontrivial task. It forces us to reassess basic notions of the nonlinguistic context and rhetorical relations as well as models of discourse structure, evolution, and interpretation. Our paper addresses the conceptual and technical revisions that these types of interaction demand.
```
@article{hunter:etal:2018,
author = {Julia Hunter and Nicholas Asher and Alex Lascarides},
year = {2018},
title = {A Formal Semantics for Situated Conversation}, 
journal = {Semantics and Pragmatics},
doi = {10.3765/sp.11.10}
}
```
Alahverdzhieva, K., A. Lascarides and D. Flickinger (2017) Aligning Speech and Co-speech Gesture in a Constraint-based Grammar, Journal of Language Modelling, 55(3), pp421--464.

This paper concerns the form-meaning mapping of communicative actions consisting of speech and improvised co-speech gestures. Based on the findings of previous cognitive and computational approaches, we advance a new theory in which this form-meaning mapping is analysed in a constraint-based grammar. Motivated by observations in naturally occurring examples, we propose several construction rules, which use linguistic form, gesture form and their relative timing to constrain the derivation of a single speech-gesture syntax tree, from which a meaning representation can be composed via standard methods for semantic composition. The paper further reports on implementing these speech-gesture construction rules within the English Resource Grammar (Flickinger, 2000). Since gestural form often underspecifies its meaning, the logical formulae that are composed via syntax are underspecified so that current models of the semantics/pragmatics interface support the range of possible interpretations of the speech-gesture act in its context of use.
```
@article{alahverdzhieva:etal:2017,
author = {K.\ Alahverdzhieva and A.\ Lascarides and D.\ Flickinger},
year = {2017},
title = {Aligning Speech and Co-Speech Gesture in a Constraint Based Grammar}, 
journal = {Journal of Language Modelling},
volume = {55},
number = {3},
pages = {421--464},
publisher = {Insitute of Computer Science of the Polish Academy of Sciences}
}
```
Cadilhac, A., N. Asher, A. Lascarides and F. Benamara (2015) Preference change, in Journal of Logic, Language and Information, 24(3), pp267--288.

Most models of rational action assume that all possible states and actions are pre-defined and that preferences change only when beliefs do. But several decision and game problems lack these features, calling for a dynamic model of preferences: preferences can change when unforeseen possibilities come to light or when there is no specifiable or measurable change in belief. We propose a formally precise dynamic model of preferences that extends an existing static model (Boutilier et al, 2004). Our axioms for updating preferences preserve consistency while minimising change, like Hansson's (1995). But unlike prior models of preference change, ours supports default reasoning with partial preference information, which is essential to handle decision problems where the decision tree isn't surveyable. We also show that our model avoids problems for other models of preference change discussed in Spohn (2009).
```
@article{cadilhac:etal:2015,
author = {A.\ Cadilhac and N.\ Asher and A.\ Lascarides and F.\ Benamara},
year = {2015},
title = {Preference Change},
journal = {Journal of Logic, Language and Information},
volume = {24},
number = {3},
pages = {267--288},
doi = {10.1007/s10849-015-9221-8}
}
```
Asher, N. and A. Lascarides (2013) Strategic Conversation, Semantics and Pragmatics, 6, 2:1--62.

Models of conversation that rely on a strong notion of cooperation don't handle dialogues where the agents' goals conflict; for instance, courtroom cross examination and political debate. We provide a game-theoretic framework in which both cooperative and non-cooperative conversation can be analysed, as well as a proof theoretic analysis, in which we prove a correspondence between a situation where the agents' preferences normally align and Gricean principles of cooperative conversation (e.g., Sincerity). We introduce the notion of safety in discourse interpretation: our logic provides the means to test whether an implicature in non-cooperative conversation can be treated as a matter of public record.
```
@article{asher:lascarides:2013,
author = {N.\ Asher and A.\ Lascarides},
year = {2013},
title = {Strategic Conversation},
journal = {Semantics and Pragmatics},
volume = {6},
number = {2},
pages = {2:1--:62},
doi = {dx.doi.org/10.3765/sp.6.2}
}
```
Asher, N. and A. Lascarides (2011) Reasoning Dynamically about What One Says, Synthese, 183(1), pp5--31, Kluwer Academic Press.

In this paper we make SDRT's glue logic for computing logical form dynamic. This allows us to model a dialogue agent's understanding of what the update of the semantic representation of the dialogue would be after his next contribution, including the effects of the rhetorical moves that he is contemplating performing next. This is a pre-requisite for developing a model of how agents reason about what to say next. We make the glue logic dynamic by using a dynamic public announcement logic (PAL). We extend PAL with a particular variety of default reasoning suited to reasoning about discourse---this default reasoning being an essential component of inferring the pragmatic effects of one's dialogue moves. We add to the PAL language a new type of announcement, known as ceteris paribus announcement, and this is used to model how an agent anticipates the (default) pragmatic effects of his next dialogue move. Our extended PAL validates certain intuitive patterns of default inference that existing PALs for practical reasoning do not. We prove that the dynamic glue logic has a PSPACE validity problem, and as such is no more complex than PAL with multiple modal operators.
```
@article{asher:lascarides:inpress,
author = {Nicholas Asher and Alex Lascarides},
year = {2011},
title = {Reasoning Dynamically about What One Says},
journal = {Synthese},
volume = {183},
number = {1},
pages = {5--31},
publisher = {Kluwer Academic Publishers}
}
```
Lascarides, A. and M. Stone (2009) Discourse Coherence and Gesture Interpretation, Gesture, 9(2), pp147--180, John Benjamins Publishing Company.

In face-to-face conversation, communicators orchestrate multimodal contributions that meaningfully combine the linguistic resources of spoken language and the visuo-spatial affordances of gesture. In this paper, we characterise this meaningful combination in terms of the coherence of gesture and speech. Descriptive analyses illustrate the diverse ways gesture interpretation can supplement and extend the interpretation of prior gestures and accompanying speech. We draw certain parallels with the inventory of coherence relations found in discourse between successive sentences. In both domains, we suggest, interlocutors make sense of multiple communicative actions in combination by using these coherence relations to link the actions' interpretations into an intelligible whole. Descriptive analyses also emphasise the improvisation of gesture; the abstraction and generality of meaning in gesture allows communicators to interpret gestures in open-ended ways in new utterances and contexts. We draw certain parallels with interlocutors' reasoning about underspecified linguistic meanings in discourse. In both domains, we suggest, coherence relations facilitate meaning-making by resolving the meaning of each communicative act through constrained inference over information made salient in the prior discourse. Our approach to gesture interpretation lays the groundwork for formal and computational models that go beyond previous approaches based on compositional syntax and semantics, in better accounting for the flexibility and the constraints found in the interpretation of speech and gesture in conversation. At the same time, it shows that gesture provides an important source of evidence to sharpen the general theory of coherence in communication.
```
@article{lascarides:stone:2009,
author =  	 {Alex Lascarides and Matthew Stone},
  title = 	 {Discourse Coherence and Gesture Interpretation},
  year = 	 {2009},
journal = {Gesture},
volume = {9},
number = {2},
pages = {147--180},
publisher = {John Benjamins Publishing Company}
}
```
Lascarides, A. and M. Stone (2009) A Formal Semantic Analaysis of Gesture, Journal of Semantics, 26(4), pp393--449, Oxford University Press.

The gestures that speakers use in tandem with speech include not only conventionalised actions with identifiable meanings (so called narrow gloss gestures or emblems) but also productive iconic and deictic gestures whose form and meanings seem largely improvised in context. In this paper, we bridge the descriptive tradition with formal models of reference and discourse structure so as to articulate an approach to the interpretation of these productive gestures. Our model captures gestures' partial and incomplete meanings as derived from form, and accounts for the more specific interpretations they derive in context. Our work emphasises the commonality of the pragmatic mechanisms for interpreting both language and gesture, and the place of formal methods in discovering the principles and knowledge that those mechanisms rely on.
```
@article{lascarides:stone:2009,
author = {Alex Lascarides and Matthew Stone},
year = {2009},
title = {A Formal Semantic Analysis of Gesture},
journal = {Journal of Semantics}
volume = {26},
number = {4},
pages = {393--449},
publisher = {Oxford University Press}
}
```
Lascarides, A. and N. Asher (2009) Agreement, Disputes and Commitments in Dialogue, Journal of Semantics, 26(2), pp109--158.

This paper provides a logically precise analysis of agreement and disputes in dialogue. The semantics distinguishes among the public commitments of each dialogue agent, including commitments to relational speech acts or rhetorical relations (e.g., Narration, Explanation, Correction). Agreement is defined to be the shared entailments of the agents' public commitments. We show that this makes precise predictions about implicit agreement. The theory also provides a consistent interpretation of disputes and models what content is agreed upon when a dispute has taken place.
```
@article{lascarides:asher:2009,
author  = {Alex Lascarides and Nicholas Asher},
year = {2009},
title = {Agreement, Disputes and Commitments in Dialogue},
journal = {Journal of Semantics},
volume = {26},
number = {2},
pages = {109--158},
publisher = {Oxford University Press}
}
```
Sporleder, C. and A. Lascarides (2008) Using Automatically Labelled Examples to Classify Rhetorical Relations: An Assessment, Natural Language Engineering, 14(3) pp369--416.

Being able to identify the rhetorical relations, such as contrast or explanation, that hold between spans of text is important for many natural language processing (NLP) applications. Using machine learning to obtain a classifier which can distinguish between different relations typically depends on the availability of manually labelled training data, which is very time-consuming to create. However, rhetorical relations are sometimes lexically marked, i.e., signalled by discourse markers (e.g., because, but, consequently etc.), and it has been suggested (Marcu and Echihabi, 2002) that the presence of these cues in some examples can be exploited to label them automatically with the corresponding relation. The discourse markers are then removed and the automatically labelled data are used to train a classifier to determine relations even when no discourse marker is present (based on other linguistic cues such as word co-occurrences).

In this paper, we investigate empirically how feasible this approach is. In particular, we test whether automatically labelled, lexically marked examples are really suitable training material for classifiers that are then applied to unmarked examples (i.e., examples which naturally occur without a discourse marker). We also explore how training on automatically labelled examples compares to training on manually labelled, unmarked examples. Our results suggest that automatically labelled data are of very limited use for classifying rhetorical relations in unmarked examples.
```
@article{sporleder:lascarides:2008,
author = {Caroline Sporleder and Alex Lascarides},
year = {2008},
title = {Using Automatically Labelled Examples to Classify Rhetorical
Relations: A Critical Assessment},
journal = {Natural Language Engineering},
volume = {14},
number = {3},
pages = {369--416},
publisher = {Cambridge University Press}
}
```
Lapata, M. and A. Lascarides (2006) Learning Sentence-internal Temporal Relations, Journal of Artificial Intelligence Research, 27, pp85--117.

In this paper we propose a data intensive approach for inferring sentence-internal temporal relations. Temporal inference is relevant for practical NLP applications which either extract or synthesize temporal information (e.g., summarisation, question answering). Our method bypasses the need for manual coding by exploiting the presence of markers like "after", which overtly signal a temporal relation. We first show that models trained on main and subordinate clauses connected with a temporal marker achieve good performance on a pseudo-disambiguation task simulating temporal inference (during testing the temporal marker is treated as unseen and the models must select the right marker from a set of possible candidates). Secondly, we assess whether the proposed approach holds promise for the semi-automatic creation of temporal information. Specifically, we use a model trained on noisy and approximate data (i.e., main and subordinate clauses) to predict intra-sentential relations present in TimeBank, a corpus containing rich temporal annotations. Our experiments compare and contrast several probabilistic models differing in their feature space, linguistic assumptions and data requirements. We evaluate performance against gold standard corpora and also against human subjects.
```
@article{lapata:lascarides:2006,
author = {Mirella Lapata and Alex Lascarides},
year = {2006},
title = {Learning Sentence-internal Temporal Relations},
journal = {Journal of Artificial Intelligence Research}
volume = {27},
pages = {85--117}
}
```
Grover, C., M. Lapata and A. Lascarides (2005) A Comparison of Parsing Technologies for the Biomedical Domain, Jounal of Natural Language Engineering, 11(1), pp25--65.

This paper reports on a number of experiments which are designed to investigate the extent to which current NLP resources are able to syntactically and semantically analyse biomedical text. We address two tasks: parsing a real corpus with a hand-built wide-coverage grammar, producing both syntactic analyses and logical forms and automatically computing the interpretation of compound nouns where the head is a nominalisation (e.g., hospital arrival means an arrival at hospital, while patient arrival means an arrival of a patient). For the former task we demonstrate that flexible and yet constrained `pre-processing' techniques are crucial to success: these enable us to use part-of-speech tags to overcome inadequate lexical coverage, and to `package up' complex technical expressions prior to parsing so that they are blocked from creating misleading amounts of syntactic complexity. We argue that the XML-processing paradigm is ideally suited for automatically preparing the corpus for parsing. For the latter task, we compute interpretations of the compounds by exploiting surface cues and meaning paraphrases, which in turn are extracted from the parsed corpus. This provides an empirical setting in which we can compare the utility of a deep parser vs. a shallow one, exploring the trade-off between resolving attachment ambiguities on the one hand and generating errors in the parses on the other. We demonstrate that a robust and reliable model of the meaning of compound nominalisations is achievable with the aid of current broad-coverage parsers.
```
@article{grover:etal:2005,
AUTHOR = {Claire Grover and Mirella Lapata and Alex Lascarides},
YEAR = {2005},
TITLE = {A Comparison of Parsing Technology for the Biomedical Domain},
journal = {Natural Language Engineering},
volume = {11},
number = {1},
pages = {25--65}
}
```
Lapata, M. and A. Lascarides (2003) A Probabilisitic Account of Logical Metonymy, Computational Linguistics, 29(2), pp263--317.

In this paper we investigate logical metonymy, i.e., constructions involving a form of semantic type coercion, in that the semantic type of the argument of a word in syntax appears to be different from the semantic type of that argument in logical form (e.g.,enjoy the book means enjoy reading the book, and easy problem means a problem that is easy to solve). The systematic variation in the interpretation of such constructions suggest a rich and complex theory of composition on the syntax/semantics interface (e.g., Pustejovsky, 1995). But the generative devices which are used to model logical metonymy typically fail to exhaustively describe all the possible interpretations, or they don't rank those interpretations in terms of their likelihood. In view of this, we acquire the meanings of metonymic verbs and adjectives from a large corpus and propose a probabilistic model which provides a ranking on the set of possible interpretations. We identify lexical semantic information automatically by exploiting the consistent correspondences between surface syntactic cues and lexical meaning. We evaluate our results against paraphrase judgements elicited experimentally from humans, and show that the model's ranking of meanings correlates reliably with human intuitions: meanings that are found highly probable by the model are also rated as plausible by the human subjects.
```
@article{lapata:lascarides:2003,
author = {Mirella Lapata and Alex Lascarides},
year = {2003},
title = {A Probabilistic Account of Logical Metonymy},
journal = {Computational Linguistics},
volume = {29},
number = {2},
pages = {263--317},
publisher = {MIT Press}
}
```
Asher, N. and A. Lascarides (2001) Indirect Speech Acts, Synthese, 128(1--2), pp183--228, Kluwer Academic Press.

In this paper, we address several puzzles concerning speech acts, particularly indirect speech acts. We show how a formal semantic theory of discourse interpretation can be used to define speech acts and to avoid murky issues concerning the metaphysics of action. We provide a formally precise definition of indirect speech acts, including the subclass of so-called conventionalized indirect speech acts. This analysis draws heavily on parallels between phenomena at the speech act level and the lexical level. First, we argue that, just as co-predication shows that some words can behave linguistically as if they're `simultaneously' of incompatible semantic types, certain speech acts behave this way too. Secondly, as Horn and Bayer (1984) and others have suggested, both the lexicon and speech acts are subject to a principle of blocking or ``preemption by synonymy'': Conventionalised indirect speech acts can block their `paraphrases' from being interpreted as indirect speech acts, even if this interpretation is calculable from Gricean-style principles. We provide a formal model of this blocking, and compare it with existing accounts of lexical blocking.
```
@article{asher:lascarides:2001,
author = {Nicholas Asher and Alex Lascarides},
year = {2001},
title = {Indirect Speech Acts},
journal = {Synthese},
volume = {128},
number = {1--2},
pages = {183--228}
}
```
Lascarides, A. and A. Copestake (1999) Default Representation in Constraint-based Frameworks, Computational Linguistics, 25(1), pp55-105, MIT Press.

Default unification has been used in several linguistic applications. Most of them have utilised defaults at a meta-level, as part of an extended description language. We propose that allowing default unification to be a fully integrated part of a typed feature structure system requires default unification to be a binary, order independent function, so that it acquires the perspicuity and declarativity familiar from normal unification-based frameworks. Furthermore, in order to respect the behaviour of defaults, default unification should allow default reentrancies and values on more general types to be overridden by conflicting default information on more specific types. We define what we believe is the first definition of default unification to satisfy these criteria, and argue that it can improve the declarativity of existing uses of default inheritance within the lexicon (because it doesn't require one to pre-specify the order in which information is accumulated) without loss of expressivity (because it validates the overriding of general defaults by conflicting more specific ones). We also argue that some linguistic phenomena suggest that there are conventional default generalisations that persist as default beyond the lexicon, and are potentially overridden by more open-ended pragmatic reasoning. We demonstrate that our version of default unification can be used to model this. Finally, we discuss the complexity of the operation, and argue that the overhead of using it in practical systems need not be large.
```
@article{lascarides:copestake:1999,
author = {Alex Lascarides and Ann Copestake},
year = {1999},
title = {Default Representation in Constraint-based Frameworks},
journal = {Computational Linguistics},
volume = {128},
number = {1},
pages = {55--105}
}
```
Asher, N. and A. Lascarides (1998) The Semantics and Pragmatics of Presupposition, Journal of Semantics, 15, pp239-299, Oxford University Press.

In this paper, we offer a novel analysis of presuppositions, paying particular attention to the interaction between the knowledge resources that are required to interpret them. The analysis has two main features. First, we capture an analogy between presuppositions, anaphora and scope ambiguity (cf. van der Sandt, 1992), by utilising semantic underspecification (cf. Reyle, 1993). Second, resolving this underspecification requires reasoning about how the presupposition is rhetorically connected to the discourse context.

This has several consequences. First, since pragmatic information plays a role in computing the rhetorical relation, it also constrains the interpretation of presuppositions. Our account therefore goes beyond existing ones, and provides a forum for analysing problematic data, that require pragmatic reasoning. Second, binding presuppositions to the context via rhetorical links replaces accommodating them, in the sense of adding them to the context (cf. Lewis, 1979). Thus, unlike previous theories, we don't resort to interpretation mechanisms that are peculiar to presuppositions. Rather, they are handled entirely in terms of the discourse update procedure.

We formalise this approach in SDRT (Asher 1993, Lascarides and Asher 1993), and demonstrate that it provides a rich framework for interpreting presuppositions, where semantic and pragmatic constraints are integrated.
```
@article{asher:lascarides:1998,
author = {Nicholas Asher and Alex Lascarides},
year = {1998},
title = {The Semantics and Pragmatics of Presupposition},
journal  = {Journal of Semantics},
volume = {15},
number = {2},
pages = {239--299},
publisher = {Oxford University Press}
}
```
Asher, N. and A. Lascarides (1998) Bridging, Journal of Semantics, 15(1), pp83-113, Oxford University Press.

In this paper, we offer a novel analysis of bridging, paying particular attention to definite descriptions. We argue that extant theories don't do justice to the way different knowledge resources interact. In line with Hobbs (1979), we claim that the rhetorical connections between the propositions introduced in the text plays an important part. But our work is distinct from his in that we model how this source of information interacts with compositional and lexical semantics. We formalise bridging in a framework known as SDRT (Asher, 1993). We demonstrate that this provides a richer, more accurate interpretation of definite descriptions than has been offered so far.
```
@article{asher:lascarides:1998,
author = {Nicholas Asher and Alex Lascarides},
year = {1998},
title = {Bridging},
journal  = {Journal of Semantics},
volume = {15},
number = {1},
pages = {83--113},
publisher = {Oxford University Press}
}
```
Asher, N. and A. Lascarides (1998) Questions in Dialogue, Linguistics and Philosophy, 23(2), pp237-309, Kluwer Academic Publishers.

In this paper we explore how compositional semantics, discourse structure, and the cognitive states of participants all contribute to pragmatic constraints on answers to questions in dialogue. We synthesise formal semantic theories on questions and answers with techniques for discourse interpretation familiar from computational linguistics, and show how this provides richer constraints on responses in dialogue than either component can achieve alone.
```
@article{asher:lascarides:1998,
author = {Nicholas Asher and Alex Lascarides},
year = {1998},
title = {Questions in Dialogue},
journal = {Linguistics and Philosophy},
volume = {23},
number = {3},
pages = {237--309}
}
```
Lascarides, A. and A. Copestake (1998) Pragmatics and Word Meaning, Journal of Linguistics, 34(2), pp387-414, Cambridge University Press.

In this paper, we explore the interaction between lexical semantics and pragmatics. We argue that linguistic processing is informationally encapsulated and utilises relatively simple `taxonomic' lexical semantic knowledge. On this basis, defeasible lexical generalisations deliver defeasible parts of logical form. In contrast, pragmatic inference is open-ended and involves arbitrary real-world knowledge. Two axioms specify when pragmatic defaults override lexical ones. We demonstrate that modeling this interaction allows us to achieve a more refined interpretation of words in a discourse context than either the lexicon or pragmatics could do on their own.
```
@article{lascarides:copestake:1998,
  author = {Alex Lascarides and Ann Copestake},
  title = {Pragmatics and Word Meaning},
  journal = {Journal of Linguistics},
  year = {1998},
  volume = {34},
  number = {2},
  pages = {55--105},
  topic = {pragmatics;lexical-semantics;}
  }
```
Lascarides, A., E. J. Briscoe, N. Asher, and A. Copestake, (1996) Order Independent Persistent Typed Default Unification, Linguistics and Philosophy, 19(1), pp1-89, Kluwer Academic Publishers.

We define an order independent version of default unification on typed feature structures. The operation is one where default information in a feature structure typed with a more specific type, will override default information in a feature structure typed with a more general type, where specificity is defined by the subtyping relation in the type hierarchy. The operation is also able to handle feature structures where reentrancies are default. We provide a formal semantics, prove order independence and demonstrate the utility of this version of default unification in several linguistic applications. First, we show how it can be used to define multiple orthogonal default inheritance in the lexicon in a fully declarative fashion. Secondly, we show how default lexical specifications (introduced via default lexical inheritance) can be made to usefully `persist beyond the lexicon' and interact with syntagmatic rules. Finally, we outline how persistent default unification might underpin default feature propagation principles and a more restrictive and constraint-based approach to lexical rules.
```
@article{lascarides:etal:1996,
author = {Alex Lascarides and Ted Briscoe and Nicholas Asher and
  	Ann Copestake},
  title = {Order Independent and Persistent Typed Default Unification},
  journal = {Linguistics and Philosophy},
  year = {1996},
  volume = {19},
  number = {1},
  pages = {1--89},
  topic = {nm-ling;unification;default-unification;}
  }
```
Lascarides, A., A. Copestake, and E. J. Briscoe, (1996) Ambiguity and Coherence, Journal of Semantics, 13(1), pp41-65, Oxford University Press.

Several recent theories of linguistic representation treat the lexicon as a highly structured object, incorporating fairly detailed semantic information, and allowing multiple aspects of meaning to be represented in a single entry (e.g. Pustejovsky, 1991; Copestake, 1992; Copestake and Briscoe, 1995). One consequence of these approaches is that word senses cannot be thought of as discrete units which are in one-to-one correspondence with lexical entries. This has many advantages in allowing an account of systematic polysemy, but leaves the problem of accounting for effects such as zeugma and the absence of crossed readings, which have traditionally been explained in terms of multiple lexical entries, but which can also arise in examples where other criteria demand that a single entry be involved. Copestake and Briscoe (1995) claimed that these cases could be explained by discourse coherence, but did not describe how this might work. We remedy this here, by formalising a general pragmatic principle which encapsulates discourse effects on word meaning. We demonstrate how it contributes to the creation of zeugma and the non-availability of crossed readings.
```
@article{lascarides:etal:1996b,
author = {Alex Lascarides and Ann Copestake and Ted Briscoe},
year = {1996},
title = {Ambiguity and Coherence},
journal = {Journal of Semantics},
volume = {13},
number = {1},
pages = {41--65}
}
```
Asher, N. and A. Lascarides, (1995) Lexical Disambiguation in a Discourse Context, Journal of Semantics, 12(1), pp69-108, Oxford University Press.

In this paper we investigate how discourse structure affects the meanings of words, and how the meanings of words affect discourse structure. We integrate three ingredients: a theory of discourse structure called SDRT, which represents discourse in terms of rhetorical relations that glue together the propositions introduced by the text segments; an accompanying theory of discourse attachment called DICE, which computes which rhetorical relations hold between the constituents, on the basis of the reader's background information; and a formal language for specifying the lexical knowledge---both syntactic and semantic---called the LKB. Through this integration, we can model the information flow from words to discourse, and discourse to words. From words to discourse, we show how the LKB permits the rules for computing rhetorical relations in DICE to be generalised and simplified, so that a single law applies to several semantically related lexical items. From discourse to words, we encode two novel heuristics for lexical disambiguation: disambiguate words so that discourse incoherence is avoided; and disambiguate words so that rhetorical connections are reinforced. These heuristics enable us to tackle several cases of lexical disambiguation, that have until now been outside the scope of theories of lexical processing.
```
@article{asher:lascarides:1995,
author = {Nicholas Asher and Alex Lascarides},
year = {1995},
title = {Lexical Disambiguation in a Discourse Context},
journal = {Journal of Semantics},
volume = {12},
number = {1},
pages = {69--108},
publisher = {Oxford University Press}
}
```
Lascarides, A. and N. Asher (1993) Temporal Interpretation, Discourse Relations and Commonsense Entailment, Linguistics and Philosophy, 16(5), pp437-493, Kluwer Academic Publishers, Dordrecht, Holland.

This paper presents a formal account of how to determine the discourse relations between propositions introduced in a text, and the relations between the events they describe. The distinct natural interpretations of texts with similar syntax are explained in terms of defeasible rules. These characterise the effects of causal knowledge and knowledge of language use on interpretation. Patterns of defeasible entailment that are supported by the logic in which the theory is expressed are shown to underly temporal interpretation.
```
@article{lascarides:asher:1993a,
author = {Alex Lascarides and Nicholas Asher},
year = {1993},
title = {Temporal Interpretation, Discourse Relations and Commonsense
                  Entailment}, 
journal = {Linguistics and Philosophy},
volume = {16},
number = {5},
pages = {437--493},
publisher = {Kluwer Academic Publishers}
}
```
Lascarides, A. and J. Oberlander, (1993) Temporal Coherence and Defeasible Knowledge, Theoretical Linguistics, 19(1), pp1--35, Walter de Gruyter, Berlin, New York.

We discuss data involving the temporal structure of connected discourse. Questions are raised about the relation between clause order in discourse and causal order in the world, and about the coherence of certain discourses. We maintain that interpretation is contextually influenced by knowledge of the world and of pragmatics, and that the role of this knowledge should be formalised via a defeasible logic. It transpires that a constrained set of reasoning patterns underlies the retrieval of certain temporal structures. Not all defeasible logics capture the set; the data help choose between candidate logics. We demonstrate that an adequate logic characterises when a text is temporally coherent, reliable and unambiguous relative to the context. We also discuss defeasible reasoning in language generation, and some consequences for the semantics-pragmatics interface.
```
@article{lascarides:oberlander:1993,
author = {Alex Lascarides and Jon Oberlander},
year = {1993},
title = {Temporal Coherence and Defeasible Knowledge},
journal = {Theoretical Linguistics},
volume = {19},
number = {1},
pages = {1--35},
publisher = {Walter de Gruyter}
}
```
Lascarides, A. (1992) Knowledge, Causality and Temporal Representation, Linguistics, 30(5), pp941-973, Walter de Gruyter, Berlin, New York.

In this paper, a formal semantic account of the simple past tense in text is offered. The contributions to the interpretation of text made by the text's syntactic structure, semantic content, aspectual classification, world knowledge of the causal relations between events, and Gricean pragmatic maxims are all represented within a single logical framework. This feature of the theory gives rise to solutions to several puzzles concerning the relation between the descriptive order of events in text and their temporal relations in interpretation.
```
@article{lascarides:1992,
author = {Alex Lascarides},
year = {1992},
title = {Knowledge, Causality and Temporal Representation},
journal = {Linguistics},
volume = {30},
number = {5},
pages = {941--973},
publisher = {Walter d Gruyter}
}
```
Lascarides, A. (1991) The Progressive and the Imperfective Paradox, synthese, 87(6), pp401-447, Kluwer Academic Publishers, Dordrecht, Holland.

Formal semantics constitutes the framework of the research presented here, and the aim is to provide a solution to the imperfective paradox; i.e. explain why “Max was running” entails “Max ran”, but “Max was running home” does not entail “Max ran home”. This paper is divided into two parts. In Part I we evaluate what I will call the Eventual Outcome Strategy for solving the imperfective paradox. This strategy is commonly used (Dowty 1979, Hinrichs 1983, Cooper 1985), and is highly intuitively motivated. I will show, however, that the formulations of the intuitions give rise to conflicts and tensions when it comes to explaining the natural language data. In Part II we offer a new approach to tackle the imperfective paradox that overcomes the problems with the Eventual Outcome Strategy.
```
@article{lascarides:1991,
author = {Alex Lascarides},
year = {1991},
title = {The Progressive and the Imperfective Paradox},
journal = {Synthese},
volume = {87},
number = {6},
pages = {401--447},
publisher = {Kluwer Academic Publishers}
}
```

Book Chapters

Bender, E. and A. Lascarides (2013) On Modeling Scope of Inflectional Negation, in P. Hofmeister and Elisabeth Norcliffe (eds.), The Core and the Periphery: Data Driven Perspectives on Syntax inspired by Ivan A.\ Sag, pages 101--124, CSLI Publications.

In this paper, we investigate the representation of negated sentences in Minimal Recursion Semantics. We begin with its treatment in the English Resource Grammar, a broad-coverage implemented HPSG, and argue that it is largely a suitable representation for English, despite possible objections. We then explore whether it is suitable for typologically different languages: namely, those that express sentential negation via inflection on the verb, particularly Turkish and Inuktitut. We find that the interaction between negation and intersective modifiers requires a change to the way in which (at least) one of them contributes to semantic composition, and we argue for adapting the analysis of intersective modifiers.

More generally, this work can be seen as a case study of universality in semantic representation in surface-oriented compositional semantics. Such representations are necessarily somewhat language-specific. Nonetheless, we still expect to see many common structures in areas of semantics such as negation. As we strive to work with a surface-oriented, compositional framework, however, we must negotiate the surface-structural differences between languages so as to achieve those common semantic structures. Thus a cross-linguistically appropriate semantic representation must not only capture meanings as they are used in different languages but also be buildable on the basis of the diverse morphosyntactic scaffolding provided by the different languages.
```
@incollection{bender:lascarides:2013,
author = {Emily Bender and Alex Lascarides},
year = {2013},
title = {On Modeling Scope of Inflectional Negation}, 
editor = {P.\ Hofmeister and E.\ Norcliffe},
booktitle = {The Core and the Periphery: Data Driven Perspectives on Syntax inspired by Ivan A.\ Sag},
pages = {101--124},
publisher = {CSLI Publications}
}
```
Lascarides, A. and N. Asher (2007) Segmented Discourse Representation Theory: Dynamic Semantics with Discourse Structure, in H. Bunt and R. Muskens (eds.) Computing Meaning: Volume 3, pp87--124, Springer.

This paper motivates and describes a dynamic semantic theory of discourse interpretation called SDRT, which uses rhetorical relations to model the semantics/pragmatics interface. We describe the syntax and dynamic semantics of the language in which logical forms are represented, a separate but related language in which semantic underspecification is expressed as partial descriptions of logical forms, and a glue logic which uses commonsense reasoning to construct logical forms, relating the semantically underspecified forms that are generated by the grammar to their pragmatically preferred interpretations. We apply the framework to some examples involving anaphora and other kinds of semantic ambiguities.
```
@incollection{lascarides:asher:2007,
author = {Alex Lascarides and Nicholas Asher},
year = {2007},
title = {Segmented Discourse 
Representation Theory: Dynamic Semantics with Discourse Structure},
editor = {H.\ Bunt and R.\ Muskens},
booktitle = {Computing Meaning: Volume 3},
publisher = {Kluwer Academic Publishers},
pages = {87--124}
}
```
Lascarides, A. and N. Asher (2004) Imperatives in Dialogue, in P. Kuehnlein, H. Rieser and H. Zeevat (eds.) The Semantics and Pragmatics of Dialogue for the New Millenium, Benjamins.

In this paper, we offer a semantic analysis of imperatives. We explore the effects of context on their interpretation, particularly on the content of the action to be performed, and whether or not the imperative is commanded. We demonstrate that by utilising a dynamic, discourse semantics which features rhetorical relations such as Narration, Elaboration and Correction, we can capture the discourse effects as a byproduct of discourse update (i.e., the dynamic construction of logical forms). We argue that this has a number of advantages over static approaches and over plan-recognition techniques for interpreting imperatives.
```
@incollection{lascarides:asher:2004,
author = {Alex Lascarides and Nicholas Asher},
title = {Imperatives in Dialogue},
year = {2004},
editor = {P.\ Kuehnlein and H.\ Rieser and H.\ Zeevat},
booktitle = {The Semantics and Pragmatics  of Dialogue for the New Millenium},
publisher = {Benjamins}
}
```
Schlangen, D. and A. Lascarides (2003) A Compositional and Constraint-Based Approach to Non-Sentential Utterances, in Muller, S. (ed.) The Proceedings of the 10th International Conference on Head-Driven Phrase Structure Grammar, pp. 380--390, CSLI Publications.

We present an approach to non-sentential utterances like B's utterance in the following dialogue:
A: Who came to the party?
B: Peter

Such utterances pose several puzzles: they convey `sentence-types' messages (propositions, questions, requests) while being of non-sentential form; and they are constrained by syntactically and semantically by the context. We address these puzzles in our approach which is compositional, since we provide a formal semantics of these utterances which is independent of context, and constraint-based because resolution is based on collecting contextual constraints.
```
@inproceedings{schlangen:lascarides:2002,
author = {D.~Schlangen and A.~Lascarides},
year = {2002},
title = {A Compositional and Constraint-Based Approach to Non-Sentential Utterances},
booktitle = {Proceedings of the 10th International Conference on Head-Driven Phrase Structure Grammar},
pages = {380--390},
publisher = {{\sc csli}}
}
```
Asher, N. and A. Lascarides (2001) The Semantics and Pragmatics of Metaphor, in Bouillon, P. and F. Busa (eds.) The Language of Word Meaning, Cambridge University Press, pp262--289..

This paper focuses on metaphor and the interpretation of metaphor in a discourse setting. There have been several accounts put forward by eminent philosophers of language---Black (1962), Hesse (1966) and Searle (1979), among others---but none of them are satisfactory. They offer a few rules for metaphoric interpretation, but many of them are redundant, and they form a list without much coherence.

Many have thought that the principles of metaphorical interpretation cannot be formally specified (e.g., Davidson, 1984). We'll attack this position with two claims. The first is that we support the view taken by Lakoff and Johnson (1980), that some aspects of metaphor are productive. We enrich this position, by demonstrating that this productivity can be captured effectively by encoding generalisations that limit metaphorical interpretation in a constraint-based framework for defining lexical semantics. Indeed from a methodological perspective, we would claim that the productive aspects of metaphor can give the linguist clues about how to represent semantic information in lexical entries.
```
@incollection{asher:lascarides:2001a,
author = {Nicholas Asher and Alex Lascarides},
year = {2001},
title = {The Semantics and Pragmatics of Metaphor},
booktitle = {The Language of Word Meaning},
editor = {P.\ Bouillon and F.\ Busa},
pages = {262--289},
publisher = {Cambridge University Press}
}
```
Oberlander, J. and A. Lascarides, (2000) Laconic Discourses and Total Eclipses: Abduction in DICE, in Harry Bunt and William Black (eds.), Abduction, Beliefs and Context: Studies in Computational Pragmatics, pp391--412, John Benjamins, London.

The purpose of this chapter is to demonstrate one particular use of abduction in the processing of natural language discourse. DICE (Discourse In Commonsense Entailment) can be used for both interpretation and generation. For interpretation, it uses defeasible deduction to compute the discourse structures and the event structures of multi-sentential text. For generation, it uses abduction to build up specifications of text from the underlying event structures. Here we demonstrate how the information flow between impicatures on the one hand and interpretation and generation on the other can be modelled, thereby showing that DICE provides a `reversible' model of the semantics/pragmatics interface.
```
@incollection{oberlander:lascarides:2000,
author = {Jon Oberlander and Alex Lascarides},
year = {2000},
title = {Laconic Discourses and Total Eclipses: Abduction in DICE},
editor = {H.\ Bunt and W.\ Black},
booktitle = {Abduction, Beliefs and Context: Studies in Computational Pragmatics},
pages = {391--412},
publisher = {John Benjamins}
}
```
Briscoe, E. J., A. Copestake, and A. Lascarides, (1995) Blocking, in St. Dizier, P. and Viegas, E. Computational Lexical Semantics, pp273--302, Cambridge University Press.

A major motivation for the introduction of default inheritance mechanisms into theories of lexical organisation has been to account for the prevalence of the family of phenomena variously described as blocking (Aronoff, 1976:43), the elsewhere condition (Kiparsky, 1973), or preemption by synonymy (Clark and Clark, 1979:798). In Copestake and Briscoe (1991) we argued that productive processes of sense extension also undergo the same process, suggesting that an integrated account of lexical semantic and morphological processes must allow for blocking. In this paper, we review extant accounts which follow from theories of lexical organisation based on default inheritance, such as Paradigmatic Morphology (Calder, 1989), {\sc datr} (Evans \& Gazdar, 1989), ELU (Russell et al., 1991, in press), Word Grammar (Hudson, 1990; Fraser and Hudson, 1992), or the LKB (Copestake 1992; this volume; Copestake et al., in press). We argue that these theories fail to capture the full complexity of even the simplest cases of blocking and sketch a more adequate framework, based on a nonmonotonic logic that incorporates more powerful mechanisms for resolving conflict among defeasible knowledge resources (Commonsense Entailment, Asher and Morreau, 1991). Finally, we explore the similarities and differences between various phenomena which have been intuitively felt to be cases of blocking within this formal framework, and discuss the manner in which such processes might interact with more general interpretative strategies during language comprehension. Our presentation is necessarily brief and rather informal; we are primarily concerned to point out the potential advantages using a more expressive default logic for remedying some of the inadequacies of current theories of lexical description.
```
@incollection{briscoe:etal:1995,
author = {Ted Briscoe and Ann Copestake and Alex Lascarides},
year = {1995},
title = {Blocking},
editor = {P.\ St Dizier and E.\ Viegas},
booktitle = {Computational Lexical Semantics},
pages = {273--302},
publisher = {Cambridge University Press}
}
```
Lascarides, A. and J. Oberlander, (1992) Abducing Temporal Discourse, in Dale, R., Hovy, E., Rosner, D. and Stock, O. (eds.) Aspects of Automated Natural Language Generation, pp167--182, Springer Verlag.

We focus on the following question: given the causal and temporal relations between events in a knowledge base, what are the ways they can be described in extended text? We argue that we want to be able to generate laconic text, where certain temporal information remains implicit but pragmatically inferrable. An algorithm for generating laconic text is proposed, interleaving abduction and nonmonotonic deduction over a formal model of pragmatic implicature. We demonstrate that the nonmonotonicity ensures that the generation of laconic text is influenced by the preceding linguistic and extra-linguistic context.
```
@incollection{lascarides:oberlander:1992,
author = {Alex Lascarides and Jon Oberlander},
year = {1992},
title = {Abducing Temporal Discourse},
booktitle = {Aspects of Automated Natural Language Generation},
editor = {R.\ Dale and E.\ Hovy and D.\ Rosner and O.\ Stock},
pages = {167--182},
publisher = {Springer Verlag}
}
```

Conference Proceedings

Miceli-Barone, A., A. Lascarides and C. Innes (2023), Dialogue-based Generation of Self-Driving Simulation Scenarios using Large Language Models, Proceedings of the Third International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics (SPLU-RoboNLP), Singapore.

Simulation is an invaluable tool for developing and evaluating controllers for self-driving cars. Current simulation frameworks are driven by highly-specialist domain specific languages, and so a natural language interface would greatly enhance usability. But there is often a gap, consisting of tacit assumptions the user is making, between a concise English utterance and the executable code that captures the user’s intent. In this paper we describe a system that addresses this issue by supporting an extended multimodal interaction: the user can follow up prior instructions with refinements or revisions, in reaction to the simulations that have been generated from their utterances so far. We use Large Language Models (LLMs) to map the user’s English utterances in this interaction into domain-specific code, and so we explore the extent to which LLMs capture the context sensitivity that’s necessary for computing the speaker’s intended message in discourse.
```
    @inproceedings{miceli-barone_etal_2023,
author = {Antonio Miceli-Barone and Craig Innes and Alex Lascarides},
year = {2023}.
title = {Dialogue-based Generation of Self-Driving Simulation
Scenarios using Large Language Models},
booktitle = {Proceedings of the Third International Combined Workshop on
Spatial Language Understanding and Grounded Communication for Robotics
 (SPLU-RoboNLP)},
address = {Singapore}
}
```
Park, J., A. Lascarides and R. Ramamoorthy (2023), Interactive Acquisition of Fine-grained Visual Concepts by Exploiting Semantics of Generic Characterizations in Discourse, Proceedings of the 15th International Conference on Computational Semantics (IWCS), Nancy, July 2023. Best paper award

Interactive Task Learning (ITL) concerns learning about unforeseen domain concepts via natural interactions with human users. The learner faces a number of significant constraints: learning should be online, incremental and few-shot, as it is expected to perform tangible belief updates right after novel words denoting unforeseen concepts are introduced. In this work, we explore a challenging symbol grounding task---discriminating among object classes that look very similar---within the constraints imposed by ITL. We demonstrate empirically that more data-efficient grounding results from exploiting the truth-conditions of the teacher's generic statements (e.g., ``Xs have attribute Z.'') and their implicatures in context (e.g., as an answer to ``How are Xs and Ys different?'', one infers Y lacks attribute Z).
```
@inproceedings{park:etal:2023,
author = {Jay Park and Alex Lascarides and Ram Ramamoorthy},
year = {2023},
title = {Interactive Acquisition of Fine-grained Visual Concepts by Exploiting
Semantics of Generic Characterizations in Discourse},
booktitle = {Proceedings of the 15th International Conference on
Computational Semantics (IWCS)},
address = {Nancy, July 2023}
}
```

Appelgren, M. and A. Lascarides (2023), Learning Manner of Execution from Partial Corrections Proceedings of the International Conference on Autonomous Agents and Multi-Agent Systems: Extended Abstract (AAMAS) London, June 2023

Some actions must be executed in different ways depending on the context. Wiping away marker requires vigorous force while almonds require gentle force. We provide a model where an agent learns which manner to execute in which context, drawing on evidence from trial and error and verbal corrections when it makes a mistake (e.g., ``no, do it gently''). The learner's initial domain model lacks the concepts denoted by the words in the teacher's feedback: both those describing the context (e.g., almonds) and those describing manner (e.g., gently). We show that discourse coherence helps the agent refine its domain model and perform the symbol grounding that's necessary for using the guidance to solve its planning problem: to perform its actions in the current context in the correct way.
```
@inproceedings{appelgren:lascarides:2023,
author = {Mattias Appelgren and Alex Lascarides},
title = {Learning Manner of Execution from Partial Corrections},
year = {2023},
booktitle = {Proceedings of the International Conference on Autonomous Agents and
Multi-Agent Systems (AAMAS)}, 
address = {London, June 2023}
}
```
Dagan, G., F. Keller and A. Lascarides (2023) Learning the Effects of Physical Actions in a Multi-modal Environment, Findings of the 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL), Croatia, May 2023.

Large Language Models (LLMs) handle physical commonsense information inadequately. As a result of being trained in a disembodied setting, LLMs often fail to predict an action's outcome in a given environment. However, predicting the effects of an action before it is executed is crucial in planning, where coherent sequences of actions are often needed to achieve a goal. Therefore, we introduce the multi-modal task of predicting the outcomes of actions solely from realistic sensory inputs (images and text). Next, we extend an LLM to model latent representations of objects to better predict action outcomes in an environment. We show that multi-modal models can capture physical commonsense when augmented with visual information. Finally, we evaluate our model's performance on novel actions and objects and find that combining modalities help models to generalize and learn physical commonsense reasoning better.
```
@inproceedings{dagan:etal:2023,
author = {Gautier Dagan and Frank Keller and Alex Lascarides},
title = {Learning the Effects of Physical Actions in a Multi-modal Environment},
year = {2023},
booktitle = {Findings of the 17th Conference of the European Chapter
of the Association for Computational Linguistics (EACL)},
address = {Croatia, May 2023}
}
```
Rubavicius, R. and A. Lascarides (2022) Interactive Symbol Grounding with Complex Referential Expressions, Proceedings of Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), Seattle, July 2022.

We present a procedure for learning to ground symbols from a sequence of stimuli consisting of an arbitrarily complex noun phrase (e.g. ``all but one green square above both red circles.”) and its designation in the visual scene. Our distinctive approach combines: a) lazy fewshot learning to relate open-class words like green and above to their visual percepts; and b) symbolic reasoning with closed-class word categories like quantifiers and negation. We use this combination to estimate new training examples for grounding symbols that occur within a noun phrase but aren’t designated by that noun phase (e.g, red in the above example), thereby potentially gaining data efficiency. We evaluate the approach in a visual reference resolution task, in which the learner starts out unaware of concepts that are part of the domain model and how they relate to visual percepts.
```
@inproceedings{rubavicius:lascarides:2022,
author = {Rimvydas Rubavicius and Alex Lascarides},
title = {Interactive Symbol Grounding with Complex Referential Expressions},
year = {2022},
booktitle = {Proceedings of the Annual Conference of the
North American Chapter of the Association for Computational Linguistics (NAACL)},  
address = {Seattle, July 2022}
}
```
Appelgren, M. and A. Lascarides (2021) Symbol Grounding and Task Learning from Imperfect Corrections, Proceedings of the Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics (SPLU-RoboNLP), ACL-IJNLP 2021.

This paper describes a method for learning from a teacher's potentially unreliable corrective feedback in an interactive task learning setting. The graphical model uses discourse coherence to jointly learn symbol grounding, domain concepts and valid plans. Our experiments show that the agent learns its domain-level task in spite of the teacher's mistakes.
```
@inproceedings{appelgren:lascarides:2021,
author = {Mattias Appelgren and Alex Lascarides},
title = {Symbol Grounding and Task Learning from Imperfect Corrections},
year = {2021},
booktitle = {Proceedings of the Second International Combined Workshop on
Spatial Language Understanding and Grounded Communication for Robotics (SPLU-RoboNLP)},  
address = {ACL-IJNLP 2021}
}
```
Hristov, Y., D. Angelov, M. Burke, A. Lascarides and S. Ramamoorthy (2019) Disengtangled Relational Representations for Explaining and Learning from Demonstration, the 3rd Conference on Robot Learning (CoRL), Osaka, Japan. Best paper runner up award.

Learning from demonstration is an effective method for human users to instruct desired robot behaviour. However, for most non-trivial tasks of practical interest, efficient learning from demonstration depends crucially on inductive bias in the chosen structure for rewards/costs and policies. We address the case where this inductive bias comes from an exchange with a human user. We propose a method in which a learning agent utilizes the information bottleneck layer of a high-parameter variational neural model, with auxiliary loss terms, in order to ground abstract concepts such as spatial relations. The concepts are referred to in natural language instructions and are manifested in the high-dimensional sensory input stream the agent receives from the world. We evaluate the properties of the latent space of the learned model in a photorealistic synthetic environment and particularly focus on examining its usability for downstream tasks. Additionally, through a series of controlled table-top manipulation experiments, we demonstrate that the learned manifold can be used to ground demonstrations as symbolic plans, which can then be executed on a PR2 robot.
```
@inproceedings{hristov:etal:2019,
author = {Yordan Hristov and Daniel Angelov and Michael Burke and Alex Lascarides and
           Subramanian Ramamoorthy},
year = {2019},
title = {Disentangled Relational Representations for Explaining and Learning from Demonstration},
booktitle = {Proceedings of the 3rd Conference on Robot Learning (CoRL)},
address = {Osaka, Japan}
}
```
Appelgren, M. and A. Lascarides (2019) Coherence, Symbol Grounding and Interactive Task Learning, Proceedings of the 23rd Workshop on the Semantics and Pragmatics of Dialogue (SEMDIAL), Queen Mary University, London.

To teach agents through natural language interaction, we need methods for updating the agent's knowledge, given a teacher's feedback. But natural language is ambiguous at many levels and so a major challenge is for the agent to disambiguate the intended message, given the signal and the context in which it's uttered. In this paper we look at how coherence relations can be used to help disambiguate the teachers' feedback and so contribute to the agent's reasoning about how to solve their domain-level task. We conduct experiments where the agent must learn to build towers that comply with a set of rules, which the agent starts out ignorant of. It is also unaware of the concepts used to express the rules. We extend a model for learning these tasks which is based on coherence and show experimentally that our extensions can improve how fast the agent learns.
```
@inproceedings{appelgren:lascarides:2019,
author = {Mattias Appelgren and Alex Lascarides},
title = {Coherence, Symbol Grounding and Interactive Task Learning},
year = {2019},
booktitle = {Proceedings of the 23rd Workshop on the Semantics and
Pragmatics of Dialogue (SEMDIAL)},
address = {Queen Mary University, London}
}
```
Innes, C. and A. Lascarides (2019) Learning Factored Markov Decision Processes with Unawareness, Proceedings of the Conference on Uncertainty in Artificial Intelligence (UAI), Tel-Aviv, Israel.

Methods for learning and planning in sequential decision problems often assume the learner is aware of all possible states and actions in advance. This assumption is sometimes untenable. In this paper, we give a method to learn factored markov decision problems from both domain exploration and expert assistance, which guarantees convergence to near-optimal behaviour, even when the agent begins unaware of factors critical to success. Our experiments show our agent learns optimal behaviour on small and large problems, and that conserving information on discovering new possibilities results in faster convergence.
```
@inproceedings{innes:lascarides:2019,
author = {C.\ Innes and A.\ Lascarides},
year = {2019},
title = {Learning Factored Markov Decision Processes with Unawareness},
booktitle = {Proceedings of the Conference on
Uncertainty in Artificial Intelligence (UAI)},
  address = {Tel-Aviv, Israel}
  }
```
Innes, C. and A. Lascarides (2019) Learning Structured Decision Problems with Unawareness, Proceedings of the 36th International Conference on Machine Learning (ICML), Long Beach, USA.

Structured models of decision making often assume an agent is aware of all possible states and actions in advance. This assumption is sometimes untenable. In this paper, we learn influence diagrams from both domain exploration and expert assertions in a way which guarantees convergence to optimal behaviour, even when the agent starts unaware of actions or belief variables that are critical to success. Our experiments show that our agent learns optimal behaviour on small and large decision problems, and that allowing an agent to conserve information upon discovering new possibilities results in faster convergence.
```
@inproceedings{innes:lascarides:2019,
author = {C.\ Innes and A.\ Lascarides},
year = {2019},
title = {Learning Structured Decision Problems with Unawareness},
  booktitle = {Proceedings of the 36th International Conference on Machine
  Learning (ICML)},
  address = {Long Beach, USA}
  }
```
Innes, C. and A. Lascarides (2019) Learning Factored Markov Decision Processes with Unawareness, Proceedings of the International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), Montreal, Canada.

Methods for learning and planning in sequential decision problems often assume the learner is fully aware of all possible states and actions in advance. This assumption is sometimes untenable: evidence gathered via domain exploration or external advice may reveal not just information about which of the currently known states are probable, but that entirely new states or actions are possible. This paper provides a model-based method for learning factored markov decision problems from both domain exploration and contextually relevant expert corrections in a way which guarantees convergence to near-optimal behaviour, even when the agent is initially unaware of actions or belief variables that are critical to achieving success. Our experiments show that our agent converges quickly on the optimal policy for both large and small decision problems. We also explore how an expert’s tolerance towards the agent’s mistakes affects the agent’s ability to achieve optimal behaviour.
```
@inproceedings{innes:lascarides:2019,
author = {C.\ Innes and A.\ Lascarides},
year = {2019},
title = {Learning Factored Markov Decision Processes with Unawareness},
  booktitle = {Proceedings of the International Conference on Autonomous
  Agents and Multi-Agent Systems (AAMAS)},
  address = {Montreal, Canada}
  }
```
Appelgren, M. and A. Lascarides (2019) Learning Plans by Acquiring Grounded Linguistic Meanings from Corrections, Proceedings of the International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), Montreal, Canada.

We motivate and describe a novel task which is modelled on interactions between apprentices and expert teachers. In the task the agent must learn to build towers which are constrained by rules. Whenever the agent performs an action which violates a rule the teacher provides verbal corrective feedback (e.g. “No, put red blocks on blue blocks”) and answers the learner’s clarification questions. The agent must learn to build rule compliant towers from these corrections and the context in which they were given. The agent starts out unaware of the constraints as well as the domain concepts in which the constraints are expressed. Therefore an agent that takes advantage of the linguistic evidence must learn the denotations of neologisms and adapt its conceptualisation of the planning domain to incorporate those denotations. We show that an agent which does utilise linguistic evidence outperforms a strong baseline which does not.
```
@inproceedings{appelgren:lascarides:2019,
author = {M.\ Appelgren and A.\ Lascarides},
year = {2019},
title = {Learning Plans by Acquiring Grounded Linguistic Meanings from Corrections},
  booktitle = {Proceedings of the International Conference on Autonomous
  Agents and Multi-Agent Systems (AAMAS)},
  address = {Montreal, Canada}
  }
```
Hristov, Y., A. Lascarides and S. Ramamoorthy (2018) Interpretable Latent Spaces for Learning from Demonstration, Proceedings of the 2nd Conference on Robot Learning (CoRL), Zurich, Switzerland.

Effective human-robot interaction, such as in robot learning from human demonstration, requires the learning agent to be able to ground abstract concepts (such as those contained within instructions) in a corresponding highdimensional sensory input stream from the world. Models such as deep neural networks, with high capacity through their large parameter spaces, can be used to compress the high-dimensional sensory data to lower dimensional representations. These low-dimensional representations facilitate symbol grounding, but may not guarantee that the representation would be human-interpretable. We propose a method which utilises the grouping of user-defined symbols and their corresponding sensory observations in order to align the learnt compressed latent representation with the semantic notions contained in the abstract labels. We demonstrate this through experiments with both simulated and real-world object data, showing that such alignment can be achieved in a process of physical symbol grounding.
```
@inproceedings{hristov:etal:2018,
author = {Yordan Hristov and Alex Lascarides and
           Subramanian Ramamoorthy},
year = {2018},
title = {Interpretable Latent Spaces for Learning from Demonstration},
booktitle = {Proceedings of the Conference on Robot Learning},
address = {Z\"urich, Switzerland}
}
```
Dobre, M. and A. Lascarides (2018) POMCP with Human Preferences in Settlers of Catan, Proceedings of the Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE), Edmonton, Canada.

We present a suite of techniques for extending the Partially Observable Monte Carlo Planning algorithm to handle complex multi-agent games. We design the planning algorithm to exploit the inherent structure of the game. When game rules naturally cluster the actions into sets called types, these can be leveraged to extract characteristics and high-level strategies from a sparse corpus of human play. Another key insight is to account for action legality both when extracting policies from game play and when these are used to inform the forward sampling method. We evaluate our algorithm against other baselines and versus ablated versions of itself in the well-known board game Settlers of Catan.
```
@inproceedings{dobre:lascarides:2018,
author = {Mihai Dobre and Alex Lascarides},
year = {2018},
title = {POMCP with Human Preferences in Settlers of Catan},
booktitle = {Proceedings of the Conference on Artificial Intelligence
	and Interactive Digital Entertainment (AIIDE)},
address = {Edmonton, Canada}
}
```
Hristov, Y, S. Penkov, A. Lascarides and S. Ramamoorthy (2017) Grounding Symbols in Multimodal Instructions, Proceedings of the ACL workshop on Language Grounding for Robotics, Vancouver, Canada.

As robots begin to cohabit with humans in semi-structured environments, the need arises to understand instructions involving rich variability---for instance, learning to ground symbols in the physical world. Realistically, this task must cope with small datasets consisting of a particular users' contextual assignment of meaning to terms. We present a method for processing a raw stream of cross-modal input---i.e., linguistic instructions, visual perception of a scene and a concurrent trace of 3D eye tracking fixations---to produce the segmentation of objects with a correspondent association to high-level concepts. To test our framework we present experiments in a table-top object manipulation scenario. Our results show our model learns the user's notion of colour and shape from a small number of physical demonstrations, generalising to identifying physical referents for novel combinations of the words.
```
@inproceedings{hristov:etal:2017,
author = {Yordan Hristov and Svetlin Penkov and Alex Lascarides and
           Subramanian Ramamoorthy},
year = {2017},
title = {Grounding Symbols in Multi-Modal Instructions},
booktitle = {Proceedings of the ACL workshop Language Grounding for
Robotics},
address = {Vancouver, Canada}
}
```
Dobre, M. and A. Lascarides (2017), Exploiting Action Categories in Learning Complex Games, Proceedings of the IEEE Conference on Intelligent Systems, London. Best paper award.

This paper presents a model for planning in a highly complex game, where certain action types are more common than others and cyclic behaviour can also easily arise. These issues are addressed by exploiting the inherent structure among the possible options to enhance the online learning algorithm: sampling during Monte Carlo Tree Search becomes a two step process, by first sampling from a distribution over the types of legal actions followed by sampling from individual actions of the chosen type. This policy drastically reduces the breadth of the rollout as well as its depth by avoiding redundant sampling behaviour. The result is a large increase in both the performance and efficiency of the model. Another contribution of this paper is assessing the benefits of a parallel implementation and afterstates in complex games. Evaluation is done via agent simulations in the board game Settlers of Catan. The resulting agent is the first based on purely online learning strategies that can handle the full set of legal actions of the game. The evaluation shows that our model outperforms previous state-of-the-art agents while taking decisions in a time threshold tolerated by human opponents.
```
@inproceedings{dobre:lascarides:2017,
author={M.\ Dobre and A.\ Lascarides},
year = {2017},
title = {Exploiting Action Categories in Learning Complex Games},
booktitle = {Proceedings of the IEEE Conference on Intelligent Systems},
address = {London}
}
```
Dobre, M. and A. Lascarides (2017), Combining a Mixture of Experts with Transfer Learning in Complex Games, Proceedings of the AAAI Spring Symposium: Learning from Observation of Humans, Stanford

We present a supervised approach for learning policies in a highly complex game from small amounts of human data consisting of state–action pairs. Our Neural Network architecture can adapt to the varying size of the set of legal actions, thus bypassing the need to hardcode the actions in the output layer or iterate over them. This makes the training more data efficient. We use synthetic data created via game simulations among AI agents to show that a mixture of experts, where each expert predicts actions in different portions of the game, improves performance. We then show that this approach applied to human data also improves performance: in particular, using transfer learning to enable one expert to learn from another enhances performance on those portions of the game for which there is relatively little training data compared to other portions. The domain chosen for evaluation is the board game Settlers of Catan.
```
@inproceedings{dobre:lascarides:2017,
author = {Mihai Dobre and Alex Lascarides},
year = {2017},
title = {Combining a Mixture of Experts with Transfer Learning in
Complex Games}, 
booktitle = {Proceedings of the AAAI Spring Symposium: Learning from
Observation of Humans},
address = {Stanford}
}
```
Keizer, S., M. Guhe, H. Cuayahuitl, I. Efstathiou, K. Engelbrecht, M. Dobre, A. Lascarides and O. Lemon (2017), Evaluating Persuasion Strategies and Deep Reinforcement Learning methods for Negotiation Dialogue Agents Proceedings of the European Chapter of the Association for Computational Linguistics (EACL), pages 480--484, Valencia, Spain.

In this paper we present a comparative evaluation of various negotiation strategies within an online version of the game ``Settlers of Catan”. The comparison is based on human subjects playing games against artificial game-playing agents (`bots’) which implement different negotiation dialogue strategies, using a chat dialogue interface to negotiate trades. Our results suggest that a negotiation strategy that uses persuasion, as well as a strategy that is trained from data using Deep Reinforcement Learning, both lead to an improved win rate against humans, compared to previous rule-based and supervised learning baseline dialogue negotiators.
```
@inproceedings{keizer:etal:2017,
author = {Simon Keizer and Markus Guhe and Heriberto Cuayahuitl and
Ioannis Efstathiou and Klaus-Peter Engelbrecht and Mihai Dobre and
Alex Lascarides and Oliver Lemon},
year = {2017},
title = {Evaluating Persuasion Strategies  and Deep Reinforcement
Learning methods for Negotiation Dialogue Agents},
booktitle = {Proceedings of the European Chapter of the Association
for Computational Linguistics (EACL)},
pages = {480--484},
address = {Valencia}
}
```
Schloeder, J. and A. Lascarides (2015) Interpreting English Pitch Countours in Context, Proceedings of the 19th Workshop on the Semantics and Pragmatics of Dialogue (SEMDIAL), pages 131--139, Gothenburg, Sweden.

This paper presents a model of how pitch contours influence the illocutionary and perlocutionary effects of utterances in conversation. Our account is grounded in several insights from the prior literature. Our distinctive contribution is to replace earlier informal claims about the implicatures arising from intonation with logical derivations: we validate inferences in the SDRT framework that resolve the partial meaning we associate with a pitch contour to different specific interpretations in different contexts.
```
@inproceedings{schloeder:lascarides:2015,
author = {Julian Schl\"oder and Alex Lascarides},
year = {2015},
title = {Interpreting English Pitch Contours in Context},
booktitle = {Proceedings of 19th Workshop on the Semantics and
Pragmatics of Dialogue},
pages = {131--139},
address = {Gothenburg, Sweden}
}
```
Dobre, M. and A. Lascarides (2015) Online Learning and Mining Human Play in Complex Games, Proceedings of the IEEE Conference on Computational Intelligence in Games (CIG), Tainan, Taiwan.

We propose a hybrid model for automatically acquiring a policy for a complex game, which combines online learning with mining knowledge from a corpus of human game play. Our hypothesis is that a player that learns its policies by combining (online) exploration with biases towards human behaviour that’s attested in a corpus of humans playing the game will outperform any agent that uses only one of the knowledge sources. During game play, the agent extracts similar moves made by players in the corpus in similar situations, and approximates their utility alongside other possible options by performing simulations from its current state. We implement and assess our model in an agent playing the complex win-lose board game Settlers of Catan, which lacks an implementation that would challenge a human expert. The results from the preliminary set of experiments illustrate the potential of such a joint model.
```
@inproceedings{dobre:lascarides:2015,
author = {M.\ Dobre and A.\ Lascarides},
year = {2015},
title = {Online Learning and Mining Human Play in Complex Games},
booktitle = {Proceedings of the IEEE Conference on
 Computational Intelligence in Games (CIG)},
address = {Tainan, Taiwan}
}
```
Hunter, J., N. Asher and A. Lascarides (2015) Integrating Non-Linguistic Events and Discourse Structure, Proceedings of the 11th International Conference on Computational Semantics (IWCS), pages 184--194, London.

Interpreting an utterance sometimes depends on the presence and nature of non-linguistic actions. In this paper, we motivate and develop a semantic model of embodied interaction in which the contribution that non-linguistic events make to the content of the interaction is dependent on their rhetorical connections to other actions, both linguistic and non-linguistic. We support our claims with concrete examples from a corpus of online chats, comparing annotations of the linguistic-only content against annotations in which non-linguistic events in the context are taken into account.
```
@inproceedings{hunter:etal:2015,
author = {J.\ Hunter and N.\ Asher and A.\ Lascarides},
year = {2015},
title = {Integrating Non-Linguistic Events into Discourse Structure},
booktitle = {Proceedings of the 11th International Conference on Computational Semantics (IWCS)},
pages = {184--194},
address = {London}
}
```
Perret, J., S. Afantenos, N. Asher and A. Lascarides (2014) Revealing Resources in Strategic Contexts, Proceedings of the 18th Workshop on the Semantics and Pragmatics of Dialogue (SEMDIAL), pages 135--144, Edinburgh.

Identifying an optimal game strategy often involves estimating the strategies of other agents, which in turn depends on hidden parts of the game state. In this paper we focus on the win-lose game The Settlers of Catan (or Settlers), in which players negotiate over limited resources. More precisely, our goal is to map each player’s utterances in such negotiations to a model of which resources they currently possess, or don’t possess. Our approach comprises three subtasks: (a) identify whether a given utterance (dialogue turn) reveals possession of a resource, or not; (b) determine the type of resource; and (c) determine the exact interval representing the quantity involved. This information can be exploited by a Settlers playing agent to identify his optimal strategy for winning.
```
@inproceedings{perret:etal:2014,
author = {Jeremy Perret and Stergos Afantenos and Nicholas Asher and Alex Lascarides}, 
year = {2014},
title = {Revealing Resources in Strategic Contexts},
booktitle = {Proceedings of the 18th Workshop on
the Semantics and Pragmatics of Dialogue (SEMDIAL)},
address = {Edinburgh},
pages = {135--144}
}
```
Guhe, M. and A. Lascarides (2014) Persuasion in Complex Games, Proceedings of the 18th Workshop on the Semantics and Pragmatics of Dialogue (SEMDIAL), pages 62--70, Edinburgh

We study the power of persuasion in a game where each player's own preferences over the negotiation's outcomes are dynamic and uncertain. Our empirical set up supports evaluating individual aspects of the persuasion and reaction strategies in controlled ways. We show how this general method gives rise to domain-specific conclusions, in our case for The Settlers of Catan: e.g., the less scope there is for persuading during the game, the more one must ensure one gains an immediate benefit from it beyond the desired trade.
```
@inproceedings{guhe:lascarides:2014,
author = {Markus Guhe and Alex Lascarides}, 
year = {2014},
title = {Persuasion in Complex Games},
booktitle = {Proceedings of the 18th Workshop on
the Semantics and Pragmatics of Dialogue (SEMDIAL)},
address = {Edinburgh},
pages = {62--70}
}
```
Guhe, M. and A. Lascarides (2014) The Effectiveness of Persuasion in The Settlers of Catan, Proceedings of IEEE Conference on Computational Intelligence in Games (CIG), pp. 25--32, Dortmund.

We present a method for obtaining a useful symbolic model of persuasion in a complex game, where players' preferences over the outcomes of negotiations over resources are incomplete, uncertain, and dynamic. We focus on the problem of identifying the stage in the game where successfully persuading an agent to perform a particular action will have the most impact on one's chances to win. Our approach exploits empirical data from game simulations, where the set up allows us to investigate individual aspects of the policy in suitably controlled ways. We demonstrate the effectiveness of the approach within the domain of The Settlers of Catan and present some specific lessons that can be learned for this particular game, e.g. that a tipping point in the game occurs for persuasion moves that are made when the leading player reaches 7 Victory Points.
```
@inproceedings{guhe:lascarides:2014,
author = {Markus Guhe and Alex Lascarides}, 
year = {2014},
title = {The Effectiveness of Persuasion in {\em The Settlers of Catan}},
booktitle = {Proceedings of the IEEE Conference on Computational Intelligence in Games},
pages = {25--32},
address = {Dortmund}
}
```
Guhe, M. and A. Lascarides (2014) Game Strategies in The Settlers of Catan, Proceedings of IEEE Conference on Computational Intelligence in Games (CIG), pages 201--208, Dortmund.

We present an empirical framework for testing game strategies in The Settlers of Catan, a complex win--lose game that lacks any analytic solution. This framework provides the means to change different components of an autonomous agent's strategy, and to test them in suitably controlled ways via performance metrics in game simulations and via comparisons of the agent's behaviours with those exhibited in a corpus of humans playing the game. We provide changes to the game strategy that not only improve the agent's strength, but corpus analysis shows that they also bring the agent closer to a model of human players.
```
@inproceedings{guhe:lascarides:2014,
author = {Markus Guhe and Alex Lascarides}, 
year = {2014},
title = {Game Strategies in {\em The Settlers of Catan}},
booktitle = {Proceedings of the IEEE Conference on Computational Intelligence in Games},
pages = {201--208},
address = {Dortmund}
}
```
Guhe, M., A. Lascarides, K. O'Connor and V. Rieser (2013) Effects of Belief and Memory on Strategic Negotiation, Proceedings of the 17th Workshop on Semantics and Pragmatics of Dialogue (DialDam), Amsterdam.

We present an empirical framework for testing negotiation strategies in a complex win--lose game that lacks any analytic solution. We explore how different belief and memory models affect trading and win rates. We show that cognitive limitations can be compensated for by being an `optimistic' negotiator: make your desired trade offer, regardless of your beliefs about how opponents will react. In contrast, agents with good cognitive abilities can win with fewer but more effective offers. Corpus analysis shows human negotiators are somewhere in between, suggesting that they compensate for deficient memory and belief when necessary.
```
@inproceedings{guhe:etal:2013,
author = {Markus Guhe and Alex Lascarides and Kevin O'Connor and Verena Rieser}, 
year = {2013},
title = {Effects of Belief and Memory on Strategic Negotiation},
booktitle = {Proceedings of the 17th Workshop on the Semantics and Pragmatics of Dialogue (DialDam)},
address = {Amsterdam}
}
```
Cadilhac, A., N. Asher, F. Benamara and A. Lascarides (2013) Grounding Strategic Conversation: Using negotiation dialogues to predict trades in a win-lose game, Proceedings of Empirical Methods in NLP (EMNLP), Melbourne.

This paper describes a method that predicts which trades players execute during a win-lose game. Our method uses data collected from chat negotiations of the game The Settlers of Catan and exploits the conversation to construct dynamically a partial model of each player's preferences. This in turn yields equilibrium trading moves via principles from game theory. We compare our method against four baselines and show that tracking how preferences evolve through the dialogue and reasoning about equilibrium moves are both crucial to success.
```
@inproceedings{cadilhac:etal:2013,
author = {Ana\"is Cadilhac and Nicholas Asher and Farah Benemara and Alex Lascarides}, 
year = {2013},
title = {Grounding Strategic Conversation: Using negotiation dialogues to predict trades in a win-lose game},
booktitle = {Proceedings of Empirical Methods in Natural Language Processing (EMNLP)},
address = {Seattle}
pages = {357--368}
}
```
Afantenos, S., N. Asher, F. Benamara, A. Cadilhac, C. D'egremont, P. Denis, M. Guhe, S. Keizer, A. Lascarides, O. Lemon, P. Muller, S. Paul, V. Rieser and L. Vieu, (2012) Developing a corpus of strategic conversation in The Settlers of Catan, in Proceedings of the 1st Workshop on Games and NLP

We describe a dialogue model and an implemented annotation scheme for a pilot corpus of annotated online chats concerning bargaining negotiations in the game The Settlers of Catan. We will use this model and data to analyze how conversations proceed in the absence of strong forms of cooperativity, where agents have diverging motives. Here we concentrate on the description of our annotation scheme for negotiation dialogues, illustrated with our pilot data, and some perspectives for future research on the issue.
```
@inproceedings{afentenos:etal:2012}
author = {Stefanos Afantenos and Nicholas Asher and Farah Benamara and Anais Cadilhac and Cedric Degremont and Pascal Denis and Markus Guhe and Simon Keizer and Alex Lascarides and Oliver
  Lemon, Philippe Muller and Soumya Paul and Verena Rieser and Laure Vieu},
year = {2012},
title = {Developing a corpus of strategic conversation in The Settlers of
Catan},
booktitle = {Proceedings of the 1st Workshop on Games and NLP},
address = {Kanazawa, Japan}
}
```

Afantenos, S., N. Asher, F. Benamara, A. Cadilhac, C. Degremont, P. Denis, M. Guhe, S. Keizer, A. Lascarides, O. Lemon, P. Muller, S. Paul, V. Popescu, V. Rieser and L. Vieu (2012) Modelling Strategic Conversation: model, annotation design and corpus, Proceedings of 16th Workshop on the Semantics and Pragmatics of Dialogue (Seinedial), Paris.

@inproceedings{afentenos:etal:2012}
author = {Stefanos Afantenos and Nicholas Asher and Farah Benamara and Anais Cadilhac and Cedric D\'{e}gremont and Pascal Denis and Markus Guhe and Simon Keizer and Alex Lascarides and Oliver Lemon and Philippe Muller and Soumya Paul and Vladamir Popescu and Verena Rieser and Laure Vieu},
year = {2012},
title = {Modelling Strategic Conversation: model, annotation design and corpus},
booktitle = {Proceedings of the 16th Workshop on the Semantics and Pragmatics of Dialogue (Seinedial)},
address = {Paris}
}

Asher, N., A. Lascarides, O. Lemon, M. Guhe, V. Rieser, P. Muller, S. Afantenos, F. Benamara, L. Vieu, P. Denis, S. Paul, S. Keizer, and C. Degremont (2012) Modelling Strategic Conversation: the STAC Project, Proceedings of 16th Workshop on the Semantics and Pragmatics of Dialogue (Seinedial), Paris.

The STAC project will develop new, formal and robust models of non-cooperative conversation, drawing from ideas in linguistics, philosophy, computer science, and economics. The project brings a state of the art, linguistic theory of discourse interpretation together with new data-driven models of agent interaction and strategic decision making. Here we discuss the project’s linguistic underpinnings, and the conceptual and empirical challenges the project faces. We also describe the project’s current data collection activities.
```
@inproceedings{asher:etal:2012}
author = {Nicholas Asher and Alex Lascarides and Oliver Lemon and Markus Guhe and Verena Rieser and Philippe Muller and Stefanos Afantenos and Farah Benamara and Laure Vieu and Pascal Denis and Soumya Paul and Simon Keizer and Cedric D\'{e}gremont}, 
year = {2012},
title = {Modelling Strategic Conversation: the STAC Project},
booktitle = {Proceedings of the 16th Workshop on the Semantics and Pragmatics of Dialogue (Seinedial)},
address = {Paris}
}
```
Asher, N. and A. Lascarides (2012) A Cognitive Model of Conversation, Proceedings of 16th Workshop on the Semantics and Pragmatics of Dialogue (Seinedial), Paris.

This paper describes a symbolic model of rational action and decision making to support analysing dialogue. The model approximates principles of behaviour from game theory, and its proof theory makes Gricean principles of cooperativity derivable when the agents' preferences align.
```
@inproceedings{asher:lascarides:2012,
author = {Nicholas Asher and Alex Lascarides},
year = {2012},
title = {A Cognitive Model of Conversation},
booktitle = {Proceedings of the 16th Workshop on the Semantics and Pragmatics of Dialogue (Seinedial)},
address = {Paris}
}
```
Guhe, M. and A. Lascarides (2012) Trading in a Multiplayer Board Game: Towards an Analsyis of Non-Cooperative Dialogue, Proceedings of Cognitive Science, Tokyo.

We collected and analysed a small dialogue corpus of people playing The Settlers of Catan. Dialogues are trading negotiations where Gricean maxims of cooperativity often break down as players adopt conflicting intentions in their attempt to win the game. This has consequences for what information players are sharing and for the sincerity of their contributions. In this paper, we motivate and describe a two-level scheme for analysing non-cooperative dialogues, where both levels are interdependent. Each dialogue move is a move in the game (e.g., an offer to trade), and a coherent contribution to the dialogue so far, connected to a prior segment with a coherence relation, such as indirect answerhood or rejection. Parsing and generating coherence relations is computationally feasible (e.g., (Baldridge and Lascarides, 2005)), and here we'll argue that their semantics help to identify the game move, even when it is implicated rather than linguistically explicit.
```
@inproceedings{guhe:lascarides:2012,
author =  	 {Markus Guhe and Alex Lascarides},
  title = 	 {Trading in a Multiplayer Board Game: Towards an Analysis of Non-Cooperative Dialogue},
  year = 	 {2012},
booktitle = {Proceedings of Cognitive Science}, 
address = {Tokyo}
}
```
Alahverdzhieva, K. D. Flickinger and A. Lascarides (2012) Multimodal Grammar Implementation, Proceedings of NAACL-HLT, Montreal.

This paper reports on an implementation of a multimodal grammar of speech and co-speech gesture within the LKB/PET grammar engineering environment. The implementation extends the English Resource Grammar (ERG, Flickinger (2000)) with HPSG types and rules that capture the form of the linguistic signal, the form of the gestural signal and their relative timing to constrain the meaning of the multimodal action. The grammar yields a single parse tree that integrates the spoken and gestural modality thereby drawing on standard semantic composition techniques to derive the multimodal meaning representation. Using the current machinery, the main challenge for the grammar engineer is the non-linear input: the modalities can overlap temporally. We capture this by identical speech and gesture token edges. Further, the semantic contribution of gestures is encoded by lexical rules transforming a speech phrase into a multimodal entity of conjoined spoken and gestural semantics.
```
@inproceedings{alahverdzhieva:etal:2012,
author =  	 {Katya Alahverdzhieva and Dan Flickinger and Alex Lascarides},
  title = 	 {Multimodal Grammar Implementation},
  year = 	 {2012},
booktitle = {Proceedings of Annual Meeting of the North American Chapter of the Association for Computational Linguistics and Human Language Technologies (NAACL-HLT 2012)}, 
address = {Montreal}
}
```
Alahverdzhieva, K. and A. Lascarides (2011) An HPSG Approach to Synchronous Deixis and Speech, in Stefan Muller (ed.), Proceedings of the 18th International Conference on Head-Driven Phase Structure Grammar (HPSG), pages 6--24, Seattle.

In this paper, we present a constraint-based analysis of multimodal communicative signals consisting of deictic gestures and speech. Our approach re-uses standard devices from linguistics to map multimodal form to meaning, thereby accounting for the gestural ambiguity by means of established underspecification mechanisms. To specify this mapping, we use empirically extracted grammar construction rules which capture the conditions under which the speech-deixis signal is grammatical and semantically intended. We demonstrate that the constraint-based grammar framework of HSPG is expressive enough to produce multimodal lfs from syntax.
```
@inproceedings{alahverdzhieva:lascarides:2011,
author =  	 {Katya Alahverdzhieva and Alex Lascarides},
year = {2011},
  title = 	 {An HPSG Approach to Synchronous Speech and Deixis}, 
booktitle =  {Proceedings of the
18th International Conference on Head-Driven Phase Structure Grammar
(HPSG)},
editor = {S.\ M\"uller},
pages = {6--24},
publisher = {CSLI Publications},
address = {Seattle}
}
```
Alahverdzhieva, K. and A. Lascarides (2011) Semantic Composition of Multimdal Actions in Constraint-based Grammars, Proceedings of CID 2011, Agay-Roches Rouges, Var, France.

The use of speech-accompanying hand gestures to depict objects, to structure the discourse or to give directions is ubiquitous in face-to-face interaction. In this paper, we analyse multimodal signals consisting of speech and gesture from the perspective of semantic composition in constraint-based grammars: we elevate standard methods from linguistics to description of multimodal input so as to connect the semantics of the gestural signal to the semantics of the speech signal and to produce an integrated logical form.
```
@inproceedings{alahverdzhieva:lascarides:2011,
author =  	 {Katya Alahverdzhieva and Alex Lascarides},
  title = 	 {Semantic Composition of Multimodal Actions in Constraint-based Grammars},
  year = 	 {2011}, 
booktitle = {Proceedings of Constraints in Discourse (CID) 2011}, 
address = {Agay-Roches Rouges, Var, France}
}
```
Cadilhac, A., N. Asher, F. Benamara and A. Lascarides (2011) Commitments to Preferences in Dialogue, Proceedings of SIGDIAL 2011, Portland OR.

We propose a method for modelling how dialogue moves influence and are influenced by the agents' preferences. We extract constraints on preferences and dependencies among them, even when they are expressed indirectly, by exploiting discourse structure. Our method relies on a study of 20 dialogues chosen at random from the Verbmobil corpus. We then test the algorithms predictions against the judgements of naive annotators on 3 random unseen dialogues. The average annotator-system agreement and the average inter-annotator agreement show that our method is reliable.
```
@inproceedings{cadillac:etal:2011,
author = {Ana\"is Cadillac and Nicholas Asher and Farah Benamara and Alex Lascarides}, 
year = {2011},
title = {Commitments to Preferences in Dialogue},
booktitle = {Proceedings of the Annual SIGDIAL Meeting on Discourse and Dialogue}, 
address = {Portland, OR}
}
```
Alahverdzhieva, K. and A. Lascarides (2011) Integration of Speech and Gesture in a Multimodal Grammar, Proceedings of TALN 2011, Montpelier.

In this paper we present a constraint-based analysis of the form-meaning mapping of deictic gesture and its synchronous speech signal. Based on an empirical study of multimodal corpora, we capture generalisations about well-formed multimodal utterances that support the preferred interpretations in the final context-of-use. More precisely, we articulate a multimodal grammar whose construction rules use the prosody, syntax and semantics of speech, the form and meaning of the deictic signal, as well as the temporal performance of speech relative to the temporal performance of deixis to constrain the derivation of a single multimodal tree and to map it to a meaning representation. The contribution of our project is two-fold: it augments the existing NLP resources with annotated speech and gesture corpora, and it also provides the theoretical grammar framework where the semantic composition of an utterance results from its speech-and-deixis synchrony.
```
@inproceedings{alahverdzhieva:lascarides:2011,
author =  	 {Katya Alahverdzhieva and Alex Lascarides},
  title = 	 {Integretation of Speech and Deictic Gesture in a Multimodal Grammar},
  year = 	 {2011},
booktitle = {Proceedings of Traitement Automatique de Langues Naturelles (TALN 2011)}, 
address = {Montpelier}
}
```
Asher, N. and A. Lascarides (2010) Reasoning Dynamically about What One Says, Proceedings of the Workshop on Theories of Information Dynamics and Interaction and their Application to Dialogue, ESSLLI 2010, Copenhagen.

In this paper we make SDRT's glue logic for computing logical form dynamic. This allows a dialogue agent to anticipate what the update of the semantic representation of the dialogue would be after his next contribution, including the effects of the rhetorical moves that he is contemplating performing next. This is a pre-requisite for planning what to say. We make the glue logic dynamic by extending a dynamic public announcement logic (PAL) with the capacity to perform default reasoning---an essential component of inferring the pragmatic effects of one's dialogue moves. We add to the PAL language a new type of announcement, known as ceteris paribus announcement, and this is used to model how an agent anticipates the default consequences of his next dialogue move. Our extended PAL validates more intuitively compelling patterns of default inference than existing PALs for practical reaosning, and we demonstrate via the proof of reduction axioms that the dynamic glue logic, like its static version, remains decidable.
```
@inproceedings{asher:lascarides:2010,
author = {Nicholas Asher and Alex Lascarides},
year = {2010},
title = {Reasoning Dynamically about What One Says},
booktitle = {Proceedings of 
 the Workshop on Theories of Information
Dynamics and Interaction and their Application to Dialogue, at 
ESSLLI 2010},
address = {Copenhagen}
}
```
Stone, M. and A. Lascarides (2010) Coherence and Rationality in Grounding, Proceedings of the 14th SEMDIAL Workshop on the Semantics and Pragmatics of Dialogue, Poznan, pp51--58.

This paper analyses dialogues where understanding and agreement are problematic. We argue that pragmatic theories can account for such dialogues only by models that combine linguistic principles of discourse coherence and cognitive models of practical rationality.
```
@inproceedings{stone:lascarides:2010,
author =  	 {Matthew Stone and Alex Lascarides},
  title = 	 {Coherence and Rationality in Dialogue},
booktitle =  {Proceedings of the 14th SEMDIAL Workshop on the
Semantics and Pragmatics of Dialogue},
address = {Poznan},
pages = {51--58}
}
```
Alahverdzhieva, K. and A. Lascarides (2010) A Grammar for Language and Co-verbal Gesture, Proceedings of the 4th Conference of the International Society for Gesture Studies (ISGS), Frankfurt.

Meaning in everyday communication is conveyed by various signals including spoken utterances and spontaneous hand gestures. The literature has attested that gestures function in synchrony with speech to deliver an integrated message, or a `single thought', exhibit language-specific properties and are subject to formal semantic modeling. One of the challenges in modeling synchrony is to use the form of the verbal signal, the form of the gesture and their relative timing to produce an integrated meaning representation. We meet this challenge by exploiting well-established semantic composition rules for deriving meaning from the form of the multimodal action. So, while the existing grammars (HPSG, LFG, CCG) produce semantic representations for unimodal input, we argue that any formalisation of language should fit into the multimodal perspective of synchronising language and co-verbal gesture. We will further show that any formalism that interfaces syntax/semantics and prosody is well-suited for regimenting synchrony and its effects on multimodal meaning, regardless of whether the surface syntactic structure is isomorphic to prosodic structure (e.g., CCG) or not (e.g., HPSG, LFG).
```
@inproceedings{alahverdzhieva:lascarides:2010,
author =  	 {Katya Alahverdzhieva and Alex Lascarides},
  title = 	 {A Grammar for Language and Co-Verbal Gesture},
  year = 	 {2010},
booktitle = {Proceedings of the Fourth Conference of the International
Society for Gesture Studies (ISGS)},
address = {Frankfurt}
}
```
Alahverdzhieva, K. and A. Lascarides (2010) Analysing Language and Co-verbal Gesture in Constraint-based Grammars, in Stefan Muller (ed.), Proceedings of the 17th International Conference on Head-Driven Phase Structure Grammar (HPSG), pp5--25, Paris.

We demonstrate that current methods for semantic composition can be extended to multimodal language so as to produce an integrated meaning representation based on the form of the verbal signal, the form of the co-verbal gesture and their relative timing. The ambiguous gestural form provides one-to-many form-meaning mappings without violating coherence in the final interpretation.
```
@inproceedings{alahverdzhieva:lascarides:2010,
author =  	 {Katya Alahverdzhieva and Alex Lascarides},
  title = 	 {Analysing Language and Co-verbal Gesture
in Constraint-based Grammars}, 
booktitle =  {Proceedings of the
17th International Conference on Head-Driven Phase Structure Grammar
(HPSG)},
editor = {S.\ M\"uller},
pages = {5--25},
address = {Paris}
}
```
Asher, N., E. Bonzon and A. Lascarides (2010) Extracting and Modelling Preferences from Dialogue, Proceedings of the International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, Dortmund.

Dialogue moves influence and are influenced by the agents' preferences. We propose a method for modelling this interaction. We motivate and describe a recursive method for calculating the preferences that are expressed, sometimes indirectly, through the speech acts performed. These yield partial CP-nets, which provide a compact and efficient method for computing how preferences influence each other. Our study of 100 dialogues in the Verbmobil corpus can be seen as a partial vindication of using CP-nets to represent preferences.
```
@inproceedings{asher:etal:2010,
author =  	 {Nicholas Asher and Elise Bonzon and Alex Lascarides},
  title = 	 {Extracting and Modelling Preferences from Dialogue},
  year = 	 {2010},
booktitle = {Proceedings of the International Conference on
Information Processing and Management of Uncertainty in
Knowledge-based Systems (IPMU)},
address = {Dortmund}
}
```
Lascarides, A. and N. Asher (2009) The Interpretation of Questions in Dialogue, in A. Riester and T. Solstad (eds.), Proceedings of Sinn und Bedeutung 13, volume 5, SinSpeC. Working Papers of the SFB 732, University of Stuttgart, pp17--30.

We propose a dynamic semantics of questions in dialogue that tracks the public commitments of each dialogue agent, including commitments to issues raised by questions.
```
@inproceedings{lascarides:asher:2009,
author = {Alex Lascarides and Nicholas Asher},
title = {The Interpretation of Questions in Dialogue},
booktitle =      {Proceedings of Sinn und Bedeutung 13},
   year =       {2009},
   editor =     {Riester, Arndt and Solstad, Torgrim},
   volume =     {5},
   series =     {SinSpeC. Working Papers of the SFB 732},
   organization = {University of Stuttgart},
pages = {17--30}
}
```
Koller, A. and A. Lascarides (2009) A Logic of Semantic Representations for Shallow Parsing, in Proceedings of the 12th Conference of the European Chapter of the Association for Computational Linguistics (EACL), pp451--459, Athens.

One way to construct semantic representations in a robust manner is to enhance shallow language processors with semantic components. Here, we provide a model theory for a semantic formalism that is designed for this, namely Robust Minimal Recursion Semantics (RMRS). We show that RMRS supports a notion of entailment that allows it to form the basis for comparing the semantic output of different parses of varying depth.
```
@inproceedings{koller:lascarides:2009,
author = {Alexander Koller and Alex Lascarides},
year = {2009},
title = {A Logic of Semantic Representations for Shallow Parsing},
booktitle = {Proceedings of the 12th Meeting of the European Chapter of the Association for Computational Linguistics},
pages = {451--460},
address = {Athens}
}
```
Lascarides, A. and N. Asher (2008) Agreement and Disputes in Dialogue, in Proceedings of the 9th SIGDIAL Workshop on Discourse and Dialogue (SIGDIAL), Columbus, pp29--36.

In this paper we define agreement in terms of shared public commitments, and implicit agreement is conditioned on the semantics of the relational speech acts (e.g., Narration, Explanation) that each agent performs. We provide a consistent interpretation of disputes, and updating a logical form with the current utterance always involves extending it and not revising it, even if the current utterance denies earlier content.
```
@inproceedings{lascarides:asher2008,
author = {Alex Lascarides and Nicholas Asher},
year = {2008},
title = {Agreement and Disputes in Dialogue},
booktitle = {Proceedings of the 9th SIGDIAL Workshop on Discourse  and
Dialogue (SIGDIAL)},
address = {Columbus, Ohio},
pages = {29--36}
}
```
Asher, N. and A. Lascarides (2008) Commitments, Beliefs and Intentions in Dialogue, in Proceedings of the 12th Workshop on the Semantics and Pragmatics of Dialogue (Londial), London, pp35--42.

We define grounding in terms of shared public commitments, and link public commitments to other, private, attitudes within a decidable dynamic logic for computing implicatures and predicting an agent's next dialogue move.
```
@inproceedings{asher:lascarides:2008,
author = {Nicholas Asher and Alex Lascarides},
year = {2008},
title = {Commitments, Beliefs and Intentions in Dialogue},
booktitle = {Proceedings of the 12th Workshop on the Semantics and Pragmatics of Dialogue (Londial)},
address = {London},
pages = {35--42}
}
```
Lascarides, A. and M. Stone (2006) Formal Semantics For Iconic Gesture, in Proceedings of the 10th Workshop on the Semantics and Pragmatics of Dialogue (Brandial), Potsdam.

We present a formal analysis of iconic coverbal gesture. Our model describes the incomplete meaning of gesture that's derivable from its form, and the pragmatic reasoning that yields a more specific interpretation. Our formalism builds on established models of discourse interpretation to capture key insights from the descriptive literature on gesture: synchronous speech and gesture express a single thought, but while the form of iconic gesture is an important clue to its interpretation, the content of gesture can be resolved only by linking it to its context.
```
@inproceedings{lascarides:stone:2006,
author = {Alex Lascarides and Matthew Stone},
year = {2006},
title = {Formal Semantics for Iconic Gesture},
booktitle = {Proceedings of the 10th Workshop on the Semantics and Pragmatics of Dialogue (Brandial)},
address = {Potsdam}
}
```
Sporleder, C. and A. Lascarides (2005) Exploiting Linguistic Cues to Classify Rhetorical Relations, in Proceedings of Recent Advances in Natural Language Processing (RANLP), Bulgaria.

We propose a method for automatically identifying rhetorical relations. We use supervised machine learning but exploit cue phrases to automatically extract and label training data. Our models draw on a variety of linguistic cues to distinguish between the relations. We show that these feature-rich models outperform the previously suggested bigram models by more than 20%, at least for small training sets. Our approach is therefore better suited to deal with relations for which it is difficult to automatically label a lot of training data because they are rarely signalled by unambiguous cue phrases (e.g., Continuation).
```
@inproceedings{sporleder:lascarides:2005,
author = {Caroline Sporleder and Alex Lascarides},
year = {2005},
title = {Exploiting Linguistic Cues to Classify Rhetorical Relations},
booktitle = {Proceedings of Recent Advances in Natural Langauge Processing (RANLP)}, 
address = {Bulgaria}
}
```
Baldridge, J. and A. Lascarides (2005) Probabilistic Head-Driven Parsing for Discourse Structure, in Proceedings of the Ninth Conference on Computational Natural Language Learning (CoNNL), Ann Arbor.

We describe a data-driven approach to building interpretable discourse structures for appointment scheduling dialogues. We represent discourse structures as headed trees and model them with probabilistic head-driven parsing techniques. We show that dialogue-based features regarding turn-taking and domain specific goals have a large positive impact on performance. Our best model achieves an f-score of 43.2% for labelled discourse relations and 67.9% for unlabelled ones, significantly beating a right-branching baseline that uses the most frequent relations.
```
@inproceedings{baldridge:lascarides:2005a,
author = {Jason Baldridge and Alex Lascarides},
year = {2005},
title = {Probabilistic Head-Driven Parsing for Discourse Structure}, 
booktitle = {Proceedings of the Ninth Conference on Computational Natural Language Learning (CoNLL)}
}
```
Baldridge, J. and A. Lascarides (2005) Annotating Discourse Structure for Robust Semantic Interpretation, Proceedings of the Sixth International Workshop on Computational Semantics (IWCS), Tilberg The Netherlands.

To date, no one has demonstrated the feasibility of automatically learning a comprehensive model of discourse interpretation that integrates seamlessly with representations fo compositional semantics for individual sentences. Current approaches are either too imprecise in terms of the semantic import they assign to discourse-level information, or too complex to represent and compute efficiently.

Our approach to practical discourse interpretation begins with a reduced form of SDRT. Because SDRT analyses are created by referencing multiple, possibly redundant information sources, a large part of an analysis is encoded via discourse segmentation and rhetorical relations alone. SDRT's semantic interpretation of this yields information about the contextually-determined content of utterances. Resolving discourse structure is thus an important step in producing the full contextually-determined information states for dialogs. However, producing full-fledged SDRT discourse structures is complex. To simplify this task, we define a tree-form of discourse structures which can be created more efficiently, both by human annotators and discourse parsers.
```
@inproceedings{baldridge:lascarides:2005,
author = {Jason Baldridge and Alex Lascarides},
year = {2005},
title = {Annotating Discourse Structures for Robust Semantic Interpretation},
booktitle = {Proceedings of the Sixth International Workshop on Computational Semantics (IWCS)},
pages = {17--29},
address = {Tilburg, The Netherlands}
}
```
Sporleder, C. and A. Lascarides (2004) Combining Hierarchical Clustering and Machine Learning to Predict High-Level Discourse Structure, Proceedings of Coling 2004, pp.43--49.

We propose a novel method to predict the inter-paragraph discourse structure of text, i.e., to infer which paragraphs are related to each other and form larger segments on a higher level. Our method c ombines a clustering algorithm wiht a model of segment ``relatedness'' acquired in a machine learning step. The model integrates information form a variety of sources, such as word co-occurrence, lexical chains, cue phrases, punctuation and tense. Our method outperforms an approach that relies on word co-occurrence alone.
```
@inproceedings{sporleder:lascarides:2004,
author = {Caroline Sporleder and Alex Lascarides},
year = {2004},
title = {Combining Hierarchical Clustering and Machine Learning to Predict High-Level Discourse Structure},
booktitle = {Proceedings of the International Conference in Computational Linguistics (COLING)},
pages = {43--49}
}
```
Lapata, M. and A. Lascarides (2004) Inferring Sentence-internal Temporal Relations, in Proceedings of the North American Chapter of the Assocation of Computational Linguistics (NAACL), pp153--160.

In this paper we propose a data intensive approach for inferring sentence-internal temporal relations, which relies on a simple probabilistic model and assumes no manual coding. We explore various combinations of features, and evaluate performance against a gold-standard corpus and human subjects performing the same task. The best model achieves 70.7% accuracy in inferring the temporal relation between two clauses and 97.4% accuracy in ordering them, assuming that the temporal relation is known.
```
@inproceedings{lapata:lascarides:2004,
author = {Mirella Lapata and Alex Lascarides},
year = {2004},
title = {Inferring Sentence-Internal Temporal Relations},
booktitle = {Proceedings of the North American Chapter of the Association for Computational Linguistics},
pages = {153--160}
}
```
Lapata, M. and A. Lascarides (2003) Detecting Novel Compounds: The Role of Distributional Evidence, in Proceedings of the 11th Conference of the European Chapter for the Association of Computational Linguistics (EACL2003), pp235--242.

Research on the discovery of terms from corpora has focused on word sequences whose recurrent occurrence in a corpus is indicative of their terminological status, and has not addressed the issue of discovering terms when data is sparse. This becomes apparent in the case of noun compounding, which is extremely productive: more than half of the candidate compounds extracted from a corpus are attested only once. We show how evidence about established (i.e.,frequent) compounds can be used to estimate features that can discriminate rare valid compounds from rare nonce terms in addition to a variety of linguistic features than can be easily gleaned from corpora without relying on parsed text.
```
@inproceedings{lapata:lascarides:2003,
author = {Mirella Lapata and Alex Lascarides},
year = {2003},
title = {Detecting Novel Compounds: The Role of Distributional Evidence}, 
booktitle = {Proceedings of the 11th Meeting of the European Chapter of the Association for Computational Linguistics},
pages = {235--242}
}
```
Schlangen, D. and A. Lascarides (2002) Resolving Fragments Using Discourse Information, Proceedings of the Sixth International Workshop on the Semantics and Pragmatics of Dialogue (EDILOG), pp. 161--168, Edinburgh, September 2002.

This paper describes an extension of RUDI, a dialogue system component for `Resolving Underspecification with Discourse Information' (Schlangen et al., 2001). The extension handles the resolution of the intended meaning of non-sentential utterances that denote propositions or questions. Some researchers have observed that there are complex syntactic, semantic and pragmatic constraints on the acceptability of such fragments, and have used this to motivate an unmodular architecture for their analysis. In contrast, our implementation is based on a clear separation of the processes of constructing compositional semantics of fragments from those for resolving their meaning in context. This is shown to have certain theoretical and practical advantages.
```
@inproceedings{schlangen:lascarides:2002,
author = {David Schlangen and Alex Lascarides},
year = {2002},
title = {Resolving Fragments using Discourse Information},
booktitle = {Proceedings of the 6th International Workshop on the Semantics and Pragmatics of Dialogue (Edilog)},
address = {Edinburgh}
}
```
Grover, C., Klein, E., Lascarides, A. and Lapata, M. (2002) XML-Based NLP for Analysing and Annotating Medical Langauge, Proceedings of the Second Workshop on NLP and XML (NLPXML-02), Coling 2002, Taipei.

We describe the use of a suite of highly flexible XML-based NLP tools in a project for processing and interpreting text in the medical domain. The main aim of the paper is to demonstrate the central role that XML mark-up and XML NLP tools have played in the analysis process and to describe the resultant annotated corpus of MedLine abstracts. In addition to the XML tools, we have succeeded in integrating a variety of non-XML `off the shelf' NLP tools into our pipelines, so that their output is added into the mark-up. We demonstrate the utility of the annotations that result in two ways. First, we investigate how they can be used to improve parse coverage of a hand-crafted grammar that generates logical forms. And second, we investigate how they contribute to automatic lexical semantic acquisition processes.
```
@inproceedings{grover:etal:2002,
author = {Claire Grover and Ewan Klein and Alex Lascarides and Mirella Lapata},
year = {2002},
title = {{\sc xml}-based {\sc nlp} Tools for Analysing and Annotating Medical Language},
booktitle = {Proceedings of the Second International Workshop on {\sc nlp} and {\sc xml}},
address = {Coling 2002, Taipei, Taiwan}
}
```
Lascarides, A. (2001) Imperatives in Dialogue, Proceedings of the 5th International Workshop on Formal Semantics and Pragmatics of Dialogue (BI-DIALOG 2001), pp1--16, Bielefeld Germany, June 2001.

In this paper, we offer a semantic analysis of imperatives. We explore the effects of context on their interpretation, particularly on the content of the action to be performed, and whether or not the imperative is commanded. We demonstrate that by utilising a dynamic, discourse semantics which features rhetorical relations such as Narration, Elaboration and Correction, we can capture the discourse effects as a byproduct of discourse update (i.e., the dynamic construction of logical forms). We argue that this has a number of advantages over static approaches and over plan-recognition techniques for interpreting imperatives.
```
@inproceedings{lascarides:2001,
author = {Alex Lascarides}, 
year = {2001},
title = {Imperatives in Dialogue},
booktitle = {Proceedings of the 5th International Workshop on the
                  Formal Semantics and Pragmatics of Dialogue
                  (Bi-Dialog)},
note = {also to appear in Peter K\"uhnlein, Hannes Rieser and Henk
Zeevat (eds.) {\em The Semantics and Pragmatics of Dialogue}, John
Benjamins},
pages = {1--16},
address = {Bielefeld}
}
```
Schlangen, D., A. Lascarides and A. Copestake, (2001) Resolving Underspecified using Discourse Information, Proceedings of the 5th International Workshop on Formal Semantics and Pragmatics of Dialogue (BI-DIALOG 2001), pp79--93, Bielefeld Germany, June 2001.

This paper describes RUDI (Resolving Underspecification with Discourse Information), a dialogue system component which computes automatically some aspects of the content of scheduling dialogues, particularly the intended denotation of the temporal expressions, the speech acts performed and the underlying goals. RUDI has a number of nice features: it is a principled approximation of a logically precise and linguistically motivated framework for representing semantics and implicatures; it has a particularly simple architecture; and it records how reasoning with a combination of goals, semantics and speech acts serves to resolve underspecification that's generated by the grammar.
```
@inproceedings{schlangen:etal:2001,
author = {David Schlangen and Alex Lascarides and Ann Copestake},
year = {2001},
title = {Resolving Underspecification Using Discourse Information},
booktitle = {Proceedings of the 5th International Workshop on Formal
                  Semantics and Pragmatics of Dialogue (Bi-Dialog)},
pages = {79--93},
note = {Also to appear in Peter K\"uhnlein, Hannes Rieser and Henk
Zeevat (eds.) {\em The Semantics and Pragmatics of Dialogue}, John
Benjamins},
address = {Bielefeld}
}
```
Copestake, A., A. Lascarides and D. Flickinger (2001) An Algebra for Semantic Construction in Constraint-based Grammars, Proceedings of the 39th Annual Meeting of the Association for Computational Linguistics (ACL/EACL 2001), pp132--139, Toulouse, France.

We develop a framework for formalizing semantic construction within grammars expressed in typed feature structure logics, including HPSG. The approach provides an alternative to the lambda calculus; it maintains much of the desirable flexibility of unification-based approaches to composition, while constraining the allowable operations in order to capture basic generalizations and improve maintainability.
```
@inproceedings{copestake:etal:2001,
author = {Ann Copestake and Alex Lascarides and Dan Flickinger},
year = {2001},
title = {An Algebra for Semantic Construction in Constraint-based
                  Grammars},
booktitle = {Proceedings of the 39th
    Annual Meeting of the Association for Computational Linguistics
    (ACL/EACL 2001)},
pages = {132--139},
address = {Toulouse}
}
```
Grover, C. and A. Lascarides (2001) XML-based Data Preparation for Robust Deep Parsing, to appear in Proceedings of the 39th Annual Meeting of the Association for Computational Linguistics (ACL/EACL 2001), pp252--259, Toulouse, France.

We describe the use of XML tokenisation, tagging and mark-up tools to prepare a corpus for parsing. Our techniques are generally applicable but here we focus on parsing Medline abstracts with the ANLT wide-coverage grammar. Hand-crafted grammars inevitably lack coverage but many coverage failures are due to inadequacies of their lexicons. We describe a method of gaining a degree of robustness by interfacing POS tag information with the existing lexicon. We also show that XML tools provide a sophisticated approach to pre-processing, helping to ameliorate the `messiness' in real language data and improve parse performance.
```
@inproceedings{grover:lascarides:2001,
author = {Claire Grover and Alex Lascarides},
title = {{\sc xml}-based Data Preparation for Robust Deep Parsing},
year = {2001},
booktitle = {Proceedings of the 39th Annual Meeting of the Association for Computational Linguistics (ACL/EACL 2001)},
pages = {252--259},
address = {Toulouse}
}
```
Lascarides, A. and Asher, N. (1999) Cognitive States, Discourse Structure and the Content of Dialogue, in Proceedings to Amstelogue 1999, Amstelogue, March 1999.

In this paper, we focus on two puzzles concerning anaphora in dialogue. On the basis of these puzzles we argue for three criteria for solving them: the representation of dialogue content must include rhetorical relations since they impose different constraints on the antecedents to anaphora; for discourse interpretation to be computable, it must be highly modular, for example the logic for discourse update and the logic for cognitive modelling must be separate; but these different modules must be allowed to interact in complex ways. We specify an account which meets these three criteria within Segmented Discourse Representation Theory (SDRT: Asher, 1993). We illustrate our approach by first using principles or rationality and cooperativity to derive principles for computing discourse content---in particular, the rhetorical connections between the propositions. We then exploit these cognitively-motivated but discourse-level axioms to account for the puzzles concerning anaphora.
```
@inproceedings{lascarides:asher:1999,
author = {Alex Lascarides and Nicholas Asher},
year = {1999},
title = {Cognitive States, Discourse Structure and the Content of
                  Dialogue}, 
booktitle = {Proceedings of the 3rd International Workshop on the
                  Formal Semantics and Pragmatics of Dialogue
                  (Amstelogue 1999)},
pages = {1--12},
address = {Amsterdam}
}
```
Copestake, A. and A. Lascarides (1997) Integrating Symbolic and Statistical Representations: The Lexicon Pragmatics Interface, Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics (ACL97), Madrid, July 7th--12th 1997, pp136--143.

We describe a formal framework for interpretation of words and compounds in a discourse context which integrates a symbolic lexicon/grammar, word-sense probabilities, and a pragmatic component. The approach is motivated by the need to handle productive word use. In this paper, we concentrate on compound nominals. We discuss the inadequacies of approaches which consider compound interpretation as either wholly lexico-grammatical or wholly pragmatic, and provide an alternative integrated account.
```
@inproceedings{copestake:lascarides:1997,
author = {Ann Copestake and Alex Lascarides},
year = {1997},
title = {Integrating Symbolic and Statistical Representations: The
                  Lexicon Pragmatics Interface},
booktitle = {Proceedings of the 35th Annual Meeting of the Association
                  for Computational Linguistics and the 8th Meeting of
                  the European Chapter for the Association for
                  Computational Linguistics (ACL97/EACL97)},
pages = {136--143},
address = {Madrid}
}
```
Asher, N. and A. Lascarides (1996) Bridging, in R. van der Sandt, R. Blutner and M. Bierwisch (eds.), From Underspecification to Interpretation, Working Papers of the Institute for Logic and Linguistics. IBM Deutschland, Heidelberg.

In this paper, we offer a novel method for processing given information in discourse, paying particular attention to definite descriptions. We argue that extant theories don't do justice to the complexity of interaction between the knowledge resources that are used. In line with Hobbs (1979), we claim that discourse structure---as defined by the rhetorical connections between the propositions introduced in the text---is an important source of knowledge for processing given information, because rhetorical relations can change the semantic content of it. We model the processing of given information as a byproduct of computing rhetorical structure in a framework known as SDRT (Asher, 1993), which formalises the interaction between discourse structure and compositional and lexical semantics in determining semantic content. We demonstrate that it provides a richer, more accurate interpretation of definite descriptions than has been offered so far.
```
@incollection{asher:lascarides:1996,
author = {Nicholas Asher and Alex Lascarides},
year = {1996},
title = {Bridging},
booktitle = {From Underspecification to Interpretation},
pubisher = {Working Papers of the Institute for Lotgic and Linguistcs},
address  = {IBM Deutschland, Heidelberg},
editor = {R.\ van der Sandt and R.\ Blutner and M.\ Bierwisch}
}
```
Lascarides, A. (1995) The Pragmatics of Word Meaning Proceedings of the AAAI Spring Symposium Series: Representation and Acquisition of Lexical Knowledge: Polysemy, Ambiguity and Generativity, pp75--80, Stanford CA, March 1995.

A parsimonious lexicon must encode generalisations (e.g., Daelemans et al 1992). One then needs to reason about when these apply. A general consensus is that an operation known as default inheritance is useful for this (Boguraev and Pustejovsky 1990, Briscoe et al 1990, Daelemans 1987, Evans and Gazdar 1989a, Flickinger 1987, and others). A frequent motivation for using it, is to capture the overriding of regularities by subregularities in a computationally efficient manner. Information need only be stated once, instead of many times in each separate word, and default inheritance ensures that words inherit the right information.

But there's a problem with this. Many lexical generalisations are of the sort where there are exceptions to the rules, which are triggered by information which resides outside the lexicon. In particular, pragmatic knowledge can trigger exceptions and default inheritance doesn't communicate properly with pragmatics to encode this.

In this paper, we'll consider three examples where this occurs: logical metonymy (e.g., enjoy the book means enjoy reading the book), adjectives (e.g., the interpretation of fast in fast car, fast motorway, fast typist etc.), and noun-verb agreement. We'll argue for a new version of default inheritance, which allows default results of lexical generalisations to persist as default beyond the lexicon. We'll show that this persistence can be exploited by the pragmatic component, to reason about when generalisations encoded in the lexicon survive in a discourse context. We'll represent the link between the lexicon and pragmatics via two axioms. These will predict the pragmatic exceptions to lexical generalisations that arise in a discourse context. We thereby explain how words are interpreted in discourse, in a way that neither the lexicon nor pragmatics could achieve on their own.
```
@inproceedings{lascarides:1995,
author = {Alex Lascarides},
year = {1995},
title = {The Pragmatics of Word Meaning},
booktitle = {Proceedings of the AAAI Spring Symposium Series: Representation and Acquisition of Lexical Knowledge: Polysemy, Ambiguity and Generativity},
pages = {75--80}
}
```
Asher, N. and A. Lascarides, (1995) Metaphor in Discourse, Proceedings of the AAAI Spring Symposium Series: Representation and Acquisition of Lexical Knowledge: Polysemy, Ambiguity and Generativity, pp3--7, Stanford CA, March 1995.

This paper focuses on metaphor and the interpretation of metaphor in a discourse setting. There have been several accounts put forward by eminent philosophers of language---Max Black, John Searle and Donald Davidson, among others---but none of them are satisfactory. They offer a few rules for metaphoric interpretation, but many of them are redundant, and they form a list without much coherence.

Many have thought that the principles of metaphorical interpretation cannot be formally specified. We'll attack this position with two claims. Our first claim is that some aspects of metaphor are productive, and this productivity can be captured by perspicuous links between generalisations that are specified in the lexicon, and general purpose circumscriptive reasoning in the pragmatic component. Indeed from a methodological perspective, we would claim the productive aspects of metaphor can give the lexicographer clues about how to represent semantic information in lexical entries.

Moreover, it is well known that domain knowledge influences metaphorical interpretation. Our second claim takes this further, and we argue that rhetorical relations---such as Elaboration, Contrast, and Parallel, among others---that connect the meanings of segments of text together, also influence this. Through studying these cases, we learn how to link lexical processing to discourse processing in a formal framework, and we give some preliminary accounts of how the link between words and discourse determine metaphor.
```
@inproceedings{asher:lascarides:1995b,
author = {Nicholas Asher and Alex Lascarides},
year = {1995},
title = {Metaphor in Discourse},
booktitle = {Proceedings of the AAAI Spring Symposium Series: Representation and Acquisition of Lexical Knowledge: Polysemy, Ambiguity and Generativivity}, 
pages = {3--7}
}
```
Lascarides, A., and A. Copestake, (1995) Pragmatics of Word Meaning, Proceedings of Semantics and Linguistic Theory (SALT5), April 1995, Austin Texas.

In this paper, we explore the interaction between lexical semantics and pragmatics. Linguistic processing is informationally encapsulated and utilises relatively simple `taxonomic' lexical semantic knowledge. On this basis, defeasible lexical generalisations deliver defeasible parts of logical form. In contrast, pragmatics is open-ended and involves arbitrary knowledge. Two axioms specify when pragmatic defaults override lexical ones. We demonstrate that modelling this interaction allows us to achieve a more refined interpretation of words in a discourse context than either the lexicon or pragmatics could do on their own.
```
@inproceedings{lascarides:copestake:1995,
  author = {Alex Lascarides and Ann Copestake},
  title = {Pragmatics and Word Meaning},
booktitle = {Proceedings of Semantics and Linguistic Theory (SALT-5)},
  year = {1995},
  }
```
Asher, N. and A. Lascarides, (1994) Intentions and Information in Discourse, in Proceedings of the 32nd Annual Meeting of the Association of Computational Linguistics (ACL94), Las Cruces, USA, June 1994, pp34--41.

This paper is about the flow of inference between communicative intentions, discourse structure and the domain during discourse processing. We augment a theory of discourse interpretation with a theory of distinct mental attitudes and reasoning about them, in order to provide an account of how the attitudes interact with reasoning about discourse structure.
```
@inproceedings{asher:lascarides:1994,
author = {Nicholas Asher and Alex Lascarides},
year = {1994},
title = {Intentions and Information in Discourse},
booktitle = {Proceedings of the 32nd Annual Meeting of the Association
                  of Computational Linguistics (ACL94)},
pages = {34--41},
address = {Las Cruces, New Mexico}
}
```
Lascarides, A. and Asher, N. (1993) A Semantics and Pragmatics for the Pluperfect, in Proceedings of the European Chapter of the Association for Computational Linguistics (EACL93), pp250--259, Utrecht, The Netherlands.

We offer a semantics and pragmatics of the pluperfect in narrative discourse. We examine in a formal model of implicature, how the reader's knowledge about the discourse, Gricean-maxims and causation contribute to the meaning of the pluperfect. By placing the analysis in a theory where the interactions among these knowledge resources can be precisely computed, we overcome some problems with previous Reichenbachian approaches.
```
@inproceedings{lascarides:asher:1993,
author = {Alex Lascarides and Nicholas Asher},
year = {1993},
title = {A Semantics and Pragmatics for the Pluperfect},
booktitle = {Proceedings of the European Chapter of the Association
                  for Computational Linguistics (EACL93)},
pages = {250--259},
address = {Utrecht, the Netherlands}
}
```
Lascarides, A. and J. Oberlander, (1993) Temporal Connectives in a Discourse Context, in Proceedings of the European Chapter of the Association for Computational Linguistics (EACL93), pp260--268, Utrecht, The Netherlands. Nominated for a best paper award.

We examine the role of temporal connectives in multi-sentence discourse. In certain contexts, sentences containing temporal connectives that are equivalent in temporal structure can fail to be equivalent in terms of discourse coherence. We account for this by offering a novel, formal mechanism for accommodating the presuppositions in temporal subordinate clauses. This mechanism encompasses both accommodation by discourse attachment and accommodation by temporal addition. As such, it offers a precise and systematic model of interactions between presupposed material, discourse context, and the reader's background knowledge. We show how the results of accommodation help to determine a discourse's coherence.
```
@inproceedings{lascarides:oberlander:1993,
author = {Alex Lascarides and Jon Oberlander},
year = {1993},
title = {Temporal Connectives in a Discourse Context},
booktitle = {Proceedings of the
European Chapter of the Association for
Computational Linguistics (EACL93)},
pages = {260--268},
address = {Utrecht}
}
```
Blackburn, P. and A. Lascarides, (1992) Sorts and Operators for Temporal Semantics, Proceedings of the Fourth Symposium on Logic and Language, Budapest, Hungary, 1992.

An essential part of natural language understanding, and hence of formal semantics, is the interpretation of temporal expressions. But the very variety of temporal phenomena---such as tense, aspect, aktionsart, temporal adverbials, and the temporal structure of extended text---has tended to result in formal semantic analyses using a wide variety of formal tools, often of a complex nature. It seems important to try and find unifying perspectives on this work, and above all, to try and gain some insight into the logical resources needed to deal with natural language temporal phenomena.

In this paper, we show how a wide variety of temporal expressions can be analysed using a simple modal language. The underlying language is well known in the AI literature: it's Halpern and Shoham's (1986) monotonic interval-based logic. However, as it stands, this language is insufficient for natural language analysis on at least two counts. The first problem is the Reference Problem: it lacks any mechanisms for temporal reference, which are essential for an adequate treatment of tense, adverbials and indexicals. The second problem is the Ontology Problem: the language doesn't reflect the wide variety of temporal ontologies, stemming from events, states and processes, which are essential for an adequate treatment of {\sc nl} aktionsart. We will show how these two defects can be removed in a simple and uniform way: sorting. Systematic use of sorting will result in simple frameworks suitable for modelling the semantics of a wide variety of natural language temporal expressions; moreover, as we shall see, the framework is strong enough to model them in a variety of ways.

This article falls into three main parts. In the first we will show that many different kinds of referential information can easily be marked in a propositional modal language by means of sorting. There are many kinds of temporal referential information; we will examine Reichenbachian reference times, indexicals and durations. In the second part of the article, we turn our attention to temporal ontology. It's a commonly held view in temporal semantics that in order to treat many temporal phenomena, one must do justice to the rich structures underlying temporal ontology. The classical work in this area is Vendler's (1967), though the ideas stretch back much earlier (Aristotle), and have been developed in many directions since (e.g., Moens and Steedman 1988, Nakhimovsky 1988, Lascarides 1991). We will see how sorting enables some of these ideas to modeled. In the third part we combine our sorted modal languages with ideas from the default logic literature.
```
@inproceedings{blackburn:lascarides:1992,
author = {Patrick Blackburn and Alex Lascarides},
year = {2002},
title = {Sorts and Operators for Temporal Semantics},
booktitle = {Proceedings of the Fourth Symposium on Logic and  Language}, 
address = {Budapest}
}
```
Lascarides, A., N. Asher, and J. Oberlander, (1992) Inferring Discourse Relations in Context, in Proceedings of the 30th Annual Meeting of the Association of Computational Linguistics (ACL92), pp1--8, Delaware USA, June 1992.

We investigate various contextual effects on text interpretation, and account for them by providing contextual constraints in a logical theory of text interpretation. On the basis of the way these constraints interact with the other knowledge sources, we draw some general conclusions about the role of domain-specific information, top-down and bot\-tom-up discourse information flow, and the usefulness of formalisation in discourse theory.
```
@inproceedings{lascarides:etal:1992,
author = {A. Lascarides and N. Asher and J. Oberlander},
year = {1992},
title = {Inferring Discourse Relations in Context},
booktitle = {Proceedings of the 30th Annual Meeting of the Association
                  for Computational Linguistics (ACL92)},
pages = {1--8},
address = {Delaware}
}
```
Oberlander, J. and A. Lascarides, (1992) Preventing False Temporal Implicatures: Interactive Defaults for Text Generation, in Proceedings of COLING92, pp721--727, Nantes France, July 1992.

Given the causal and temporal relations between events in a knowledge base, what are the ways they can be described in text?

Elsewhere, we have argued that during interpretation, the reader-hearer H must infer certain temporal information from knowledge about the world, language use and pragmatics. It is generally agreed that processes of Gricean implicature help determine the interpretation of text in context. But without a notion of logical consequence to underwrite them, the inferences---often defeasible in nature---will appear arbitrary, and unprincipled. Hence, we have explored the requirements on a formal model of temporal implicature, and outlined one possible nonmonotonic framework for discourse interpretation (Lascarides and Asher (1991), Lascarides and Oberlander (1992a)).

Here, we argue that if the writer-speaker S is to tailor text to H, then discourse generation can be informed by a similar formal model of implicature. We suggest two ways to do it: a version of Hobbs et al's (1988, 1990) Generation as Abduction; and the Interactive Defaults strategy introduced by Joshi et al (1984a, 1984b, 1986). In investigating the latter strategy, the basic goal is to determine how notions of temporal reliability, precision and coherence can be used by a nonmonotonic logic to constrain the space of possible utterances. We explore a defeasible reasoning framework in which the interactions between the relative knowledge bases of S and H helps do this. Finally, we briefly discuss limitations of the strategy: in particular, its apparent marginalisation of discourse structure.

The paper focuses very specifically on implicatures of a temporal nature. To examine the relevant examples in sufficient detail, we have had to exclude discussion of many closely related issues in the theory of discourse structure. To motivate this restriction, let us therefore consider first why we might want to generate discourses with structures which lead to temporal complexities.
```
@inproceedings{oberlander:lascarides:1992,
author  = {Jon Oberlander and Alex Lascarides},
year = {1992},
title = {Preventing False Temporal 
Implicatures: Interactive Defaults for Text Generation},
booktitle = {Proceedings of the International Conference in Computational Linguistics (COLING)},
pages = {721--727},
address = {Nantes, France}
}
```
Lascarides, A. and N. Asher, (1991) Discourse Relations and Defeasible Knowledge, in Proceedings to the 29th Annual Meeting of the Association of Computational Linguistics (ACL91), pp55--63, Berkeley USA, June 1991.

This paper presents a formal account of the temporal interpretation of text. The distinct natural interpretations of texts with similar syntax are explained in terms of defeasible rules characterising causal laws and Gricean-style pragmatic maxims. Intuitively compelling patterns of defeasible entailment that are supported by the logic in which the theory is expressed are shown to underly temporal interpretation.
```
@inproceedings{lascarides:asher:1991,
author = {A. Lascarides and N. Asher},
year = {1991},
title = {Discourse Relations and Defeasible Knowledge},
booktitle = {Proceedings of the 29th Annual Meeting of the Association
                  for Computational Linguistics (ACL91)},
pages = {55--63},
address = {Berkeley}
}
```
Oberlander, J. and A. Lascarides, (1991) Discourse Generation, Temporal Constraints and Defeasible Reasoning, in Proceedings of the AAAI Fall Symposium on Discourse Structure in Interpretation and Generation, Asilomar, California, November 1991.

Given the causal and temporal relations between events in a knowledge base, what are the ways they can be described in text?

Elsewhere, we have argued that defeasible reasoning underlies the hearer's temporal interpretation of text. Here, we argue, on the basis of the kind of temporal information that remains implicit in candidate utterances, that if the speaker is to tailor text to the hearer, then defeasible reasoning must be integrated into the generation process. We suggest two ways in which this can be done: a version of Hobbs et al's (1988, 1990) Generation as Abduction, and the Interactive Defaults Strategy introduced by Joshi et al (1984, 1986). Assuming the Interactive Defaults strategy, the basic goal is to determine how notions of temporal reliability, precision and coherence can be used by a nonmonotonic logic to constrain the space of possible utterances. We explore a defeasible reasoning framework in which the interactions between the relative knowledge bases of speakers and hearers helps do this. Finally, we briefly discuss an objection to the programme as outlined, considering whether discourse structure has been marginalised.
```
@inproceedings{oberlander:lascarides:1992,
author  = {Jon Oberlander and Alex Lascarides},
year = {1991},
title = {Discourse Generation, Temporal Constraints and Defeasible Reasoning},
booktitle = {Proceedings of the AAAI Fall Symposium on Discourse Structure in Interpretation and Generation},
\address = {Asilomar, California}
}
```