I am a Professor of Computer Science at the School of Informatics, University of Edinburgh, and a member of EdinburghNLP, the University's Natural Language Processing Group.
My overarching research goal is to develop AI systems that go beyond pattern recognition to reason, generalise, and adapt to novel situations. I am particularly interested in models that can understand complex requests, integrate information across multiple sources and modalities, plan and execute multi-step tasks, and generate responses grounded in verifiable evidence.
My research focuses on reasoning and generalisation in language models, structured representations for long-context understanding and generation, multimodal learning, and methods for making AI systems more reliable, efficient, and interpretable. Problems that I'm currently excited about include:
• Sümeyye Meryem Taşyürek (since 2026)
• Jakob Johannes Bauer (since 2025)
• Litu Ou (since 2025)
• Akash Gupta (since September 2024)
• Ashutosh Adhikari (since September 2024)
• Alex Gurung (since September 2023)
• Argyrios Papoudakis (since September 2022; co-supervised with Frank Keller)
• Danna Zheng (since September 2022; co-supervised with Jeff Pan)
• Miao Li (since 2025)
• Iñigo Alonso (since 2025)
• Louis Mahon (since 2023)
• Laura Perez-Beltrachini (since 2018)
• Irina Saparina (PhD 2026, From Pattern Matching to Intent Reasoning in Semantic Parsing)
• Agostina Calabrese (PhD 2026, Reframing Content Moderation: Guideline-Driven and Explainable Hate Speech Detection)
• Danyang Liu (PhD 2026, Character Grounding and Planning for Visual Story Generation)
• Parag Jain (PhD 2025, Explicit Context Representations for Conversational Language Understanding)
• Tom Hosking (PhD 2024, Learning Weakly Structured Representations for Text-to-Text Generation)
• Yao Fu (PhD 2024, Improving Complex Reasoning in Large Language Models)
• Tom Sherborne (PhD 2024, Modelling Cross-lingual Transfer for Semantic Parsing)
• Hao Zheng (PhD 2023, Towards Human-like Compositional Generalization with Neural Models)
• Nelly Papalampidi (PhD 2022, Structure-aware Narrative Summarization from Multiple Views)
• Yumo Xu (PhD 2022, Document Summarization with Neural Query Modeling)
• Ratish Puduppully (PhD 2022, Data-to-Text Generation with Neural Planning)
• Reinald Kim Amplayo (PhD 2022, Opinion Summarization of Multiple Reviews: Data Synthesis and Modeling)
• Rui Cai (PhD 2021, Neural Semantic Role Labeling with more and less Supervision)
• Jonathan Mallinson (PhD 2021, Universal Rewriting via Machine Translation)
• Jiangming Liu (PhD 2021, Understanding and Generating Language with Discourse Representation Structures)
• Yang Liu (PhD 2020, Neural Document Modeling and Summarization)
• Li Dong (PhD 2019, Learning Natural Language Interfaces with Neural Models)
• Stefanos Angelidis (PhD 2019, Weakly Supervised Sentiment Analysis and Opinion Extraction)
• Jianpeng Cheng (PhD 2019, The Lifecycle of Neural Semantic Parsing)
• Philip Gorinski (PhD 2018, Automatic Movie Analysis and Summarization)
• Xingxing Zhang (PhD 2017, Natural Language Generation as Sequence Learning and Beyond)
• Siva Reddy (PhD 2017, Syntax-Mediated Semantic Parsing)
• Lea Frermann (PhD 2016, Bayesian Models of Category Acquisition and Meaning Development)
• Carina Silberer (PhD 2015, Learning Visually Grounded Meaning Representations)
• Ioannis Konstas (PhD 2014, Joint Models for Concept-to-text Generation)
• Trevor Fountain (PhD 2013, Modelling the Acquisition of Natural Language Categories)
• Joel Lang (PhD 2012, Unsupervised Induction of Semantic Roles)
• Yansong Feng (PhD 2011, Automatic Caption Generation for News Images)
• Neil McIntyre (PhD 2011, Learning to Tell Tales: Automatic Story Generation from Corpora)
• Jeff Mitchell (PhD 2011, Composition in Distributional Models of Semantics)
• Sam Brody (PhD 2009, Closing the Gap in WSD: Supervised results with Unsupervised Methods)
• James Clarke (PhD 2008, Global Inference for Sentence Compression: An Integer Linear Programming Approach)
• Sebastian Padó (PhD 2007, Cross-Lingual Annotation Projection Models for Role-Semantic Information)
This page contains publications for the past three years. For older papers, please consult my Google Scholar Profile and/or the ACL anthology.
Lightweight Latent Reasoning for Narrative Tasks
Alexander Gurung, Esmeralda S. Whitammer, Mirella Lapata.
TACL (2026)
Paper
Code
Blog
Generating Visual Stories with Grounded and Coreferent Characters
Danyang Liu, Mirella Lapata, Frank Keller.
TACL (2026)
Paper
Meta-Adaptive Prompt Distillation for Few-Shot Visual Question Answering
Akash Gupta, Amos Storkey, Mirella Lapata.
ICLR (2026)
Paper
Code
TABLET: A Large-Scale Dataset for Robust Visual Table Understanding
Iñigo Alonso, Imanol Miranda, Eneko Agirre, Mirella Lapata.
ICLR (2026)
Paper
Code
Think Before you Write: QA-Guided Reasoning for Character Descriptions in Books
Argyrios Papoudakis, Mirella Lapata, Frank Keller.
ACL Findings (2026)
Paper
Code
Long-Context Reasoning Through Proxy-Based Chain-of-Thought Tuning
Miao Li, Irina Saparina, Alexander Gurung, Mirella Lapata.
ACL (2026)
Paper
Code
Action-Level Credit Assignment for Retrieval-Augmented Reasoning
Laura Perez-Beltrachini, Mirella Lapata.
EMNLP Findings (2026), to appear
Code
GraphLit: Learning Text-Enriched Dynamic Character Network Representations for Literary Study
Gaspard Michel, Elena V. Epure, Romain Hennequin, Christophe Cerisara, Mirella Lapata.
EMNLP (2026), to appear
Paper
Code
Storyline Trees: Hierarchical Representations for Long-Form Narratives
Litu Ou, Mirella Lapata.
EMNLP Findings (2026), to appear
Paper
Code
Multimodal Latent Reasoning via Predictive Embeddings
Ashutosh Adhikari, Mirella Lapata.
COLM (2026)
Paper
Reasoning about Intent for Ambiguous Requests
Irina Saparina, Mirella Lapata.
COLM (2026)
Paper
Code
Learning Steerable Clarification Policies with Collaborative Self-play
Jonathan Berant, Maximillian Chen, Adam Fisch, Reza Aghajani, Fantine Huot, Mirella Lapata, Jacob Eisenstein.
COLM (2026)
Paper
MT-PingEval: Evaluating Multi-Turn Collaboration with Private Information Games
Jacob Eisenstein, Fantine Huot, Adam Fisch, Jonathan Berant, Mirella Lapata.
COLM (2026)
Paper
REFRAMED: Towards Realistic Audio Description Generation for Movies
Igor Sterner, Mirella Lapata, Alex Lascarides, Frank Keller.
COLM (2026)
Paper
Dataset
Uncertainty Quantification in Retrieval Augmented Question Answering
Laura Perez-Beltrachini, Mirella Lapata.
TMLR (2025)
Paper
Explanatory Summarization with Discourse-Driven Planning
Dongqi Liu, Xi Yu, Vera Demberg, Mirella Lapata.
TACL (2025)
Paper
DOLOMITES:
Domain-Specific Long-Form Methodical Tasks
Chaitanya Malaviya, Priyanka Agrawal, Kuzman Ganchev, Pranesh Srinivasan, Fantine Huot, Jonathan Berant, Mark Yatskar, Dipanjan Das, Mirella Lapata, Chris Alberti.
TACL (2025)
Paper
Code
Learning to Reason for Long-Form Story Generation
Alexander Gurung, Mirella Lapata.
COLM (2025)
Paper
Debating for Better Reasoning in Vision-Language Models
Ashutosh Adhikari, Mirella Lapata.
EMNLP Findings (2025)
Paper
Compositional Generalisation for Explainable Hate Speech Detection
Agostina Calabrese, Tom Sherborne, Björn Ross, Mirella Lapata.
EMNLP (2025)
Paper
Long-Form Information Alignment Evaluation Beyond Atomic Facts
Danna Zheng, Mirella Lapata, Jeff Z. Pan.
EMNLP (2025)
Paper
Context-Aware Hierarchical Merging for Long Document Summarization
Litu Ou, Mirella Lapata.
ACL Findings (2025)
Paper
Help Me Write a Story: Evaluating LLMs' Ability to Generate Writing Feedback
Hannah Rashkin, Elizabeth Clark, Fantine Huot, Mirella Lapata.
ACL (2025)
Paper
Decomposed Opinion Summarization with Verified Aspect-Aware Modules
Miao Li, Jey Han Lau, Eduard Hovy, Mirella Lapata.
ACL Findings (2025)
Paper
Disambiguate First, Parse Later: Generating Interpretations for Ambiguity Resolution in Semantic Parsing
Irina Saparina, Mirella Lapata.
ACL Findings (2025)
Paper
Masking in Multi-hop QA: An Analysis of How Language Models Perform with Context Permutation
Wenyu Huang, Pavlos Vougiouklis, Mirella Lapata, Jeff Z. Pan.
ACL (2025)
Paper
What Is That Talk About? A Video-to-Text Summarization Dataset for Scientific Presentations
Dongqi Liu, Chenxi Whitehouse, Xi Yu, Louis Mahon, Rohit Saxena, Zheng Zhao, Yifu Qiu, Mirella Lapata, Vera Demberg.
ACL (2025)
Paper
Agents' Room: Narrative Generation through Multi-step Collaboration
Fantine Huot, Reinald Kim Amplayo, Jennimaria Palomaki, Alice Shoshana Jakobovits, Elizabeth Clark, Mirella Lapata.
ICLR (2025)
Paper
Hierarchical Indexing
for Retrieval-Augmented Opinion Summarization
Tom Hosking, Hao Tang, Mirella Lapata.
TACL (2024)
Paper
Code
AMBROSIA: A Benchmark for Parsing Ambiguous Questions into Database Queries
Irina Saparina, Mirella Lapata.
NeurIPS (2024)
Paper
Code
Finding the Right Moment:
Human-Assisted Trailer
Creation via Task Decomposition
Pinelopi Papalampidi, Frank Keller, Mirella Lapata.
IEEE Transactions on Pattern Analysis and Machine Intelligence 46:1, 292-304
Paper
Code
Evaluating LLMs for Targeted Concept Simplification for Domain-Specific Texts
Sumit Asthana, Hannah Rashkin, Elizabeth Clark, Fantine Huot, Mirella Lapata.
EMNLP (2024)
Paper
Code
Low-Rank Adaptation for Multilingual Summarization: An Empirical Study
Chenxi Whitehouse, Fantine Huot, Jasmijn Bastings, Mostafa Dehghani,
Chu-Cheng Lin, Mirella Lapata.
NAACL Findings (2024)
Paper
Code
BookWorm: A Dataset for
Character Description and Analysis
Argyrios Papoudakis, Mirella Lapata, Frank Keller.
EMNLP Findings (2024)
Paper
Code
CHIRON: Rich Character
Representations in Long-Form Narratives
Alex Gurung, Mirella Lapata.
EMNLP Findings (2024)
Paper
Code
Less is More: Making Smaller Language Models Competent Subgraph Retrievers for Multi-hop KGQA Wenyu Huang, Guancheng Zhou, Hongru Wang, Pavlos Vougiouklis, Mirella Lapata, Jeff Z. Pan. EMNLP Findings (2024) Paper Code
Archer: A Human-Labeled Text-to-SQL Dataset with Arithmetic, Commonsense and Hypothetical Reasoning Danna Zheng, Mirella Lapata, Jeff Z. Pan. EACL (2024) Paper Code
Improving Generalization
in Semantic Parsing by Increasing Natural Language Variation
Irina Saparina, Mirella Lapata.
EACL (2024)
Paper
Code
μPLAN: Summarizing Using a Content Plan as Cross-Lingual Bridge
Fantine Huot, Joshua Maynez, Chris Alberti, Reinald Kim Amplayo,
Priyanka Agrawal, Constanza Fierro, Shashi Narayan, Mirella Lapata.
EACL (2024)
Paper
Code
PixT3: Pixel-based
Table-to-Text Generation
Iñigo Alonso, Eneko Agirre, Mirella Lapata.
ACL (2024)
Paper
Code
A Modular Approach for
Multimodal Summarization of TV Shows
Louis Mahon, Mirella Lapata.
ACL (2024)
Paper
Code
Learning to Plan and
Generate Text with Citations
Constanza Fierro, Reinald Kim Amplayo, Fantine Huot, Nicola De Cao,
Joshua Maynez, Shashi Narayan, Mirella Lapata.
ACL (2024)
Paper
Code
Explainability and Hate
Speech: Structured Explanations Make Social Media
Moderators Faster
Agostina Calabrese, Leonardo Neves, Neil Shah, Maarten Bos, Björn
Ross, Mirella Lapata, Francesco Barbieri.
ACL (2024)
Paper
Code
Little Red Riding Hood
Goes around the Globe: Crosslingual Story Planning and Generation with Large Language Models
Evgeniia Razumovskaia, Joshua Maynez, Annie Louis, Mirella Lapata, Shashi Narayan.
LREC (2024)
Paper
Code
• Science of Fundamental AI Research (SOFAIR) Lab, national AI research lab led by UCL with Cambridge, Oxford and Edinburgh, funded by UKRI (lab website), 2026-present.
• AI Hub in Generative Models, funded by UKRI and EPSRC (hub website), 2024-present.
• UKRI AI Centre for Doctoral Training in Designing Responsible Natural Language Processing, Research Training Grant, funded by UKRI (CDT website), 2024-2032.
• Teaching Machines to Reason like Humans, Turing AI Fellowship, funded by UKRI, 2021-2027.
• UKRI Centre for Doctoral Training in Natural Language Processing, Research Training Grant, funded by UKRI, 2019-2027.
• Addressing Hallucinations in Generative Language Models, Innovate UK award, funded by UKRI, 02/2024-02/2025.
• Multimodal Summarization of Creative Content, funded by Amazon, 01/2022-01/2023.
• Generating Opinion Summaries and their Explanations from User Reviews, funded by Megagon Labs, 08/2019-07/2020.
• Question-Answering with Knowledge Graph Queries, funded by Huawei, 2021-2024.
• Real-world Semantic Parsing, funded by Amazon, 2020-2021.
• Foreign Language Automated Information Retrieval (FLAIR), funded by IARPA-BAA-16-11, 10/2017-08/2021.
• Translating from Multiple Modalities into Text, Consolidator Grant (CoG), funded by the ERC, 09/2016-08/2021.
• A Unified Model of Compositional and Distributional Semantics: Theory and Applications, Research Project Grant, funded by the EPSRC, 03/2013-03/2016.
• Readers: Evaluation and Development of Reading Systems, Research Project Grant, funded by the EPSRC
• An integrated model of syntactic and semantic prediction in human language processing, Research Project Grant, funded by the EPSRC, 07/2011-02/2015.
• Global Inference for Summarization Using Integer Linear Programming, Research Project Grant, funded by the EPSRC, 01/2009-01/2012.
• Ranking Word Senses for Disambiguation: Models and Applications, Research Project Grant, funded by the EPSRC, 09/2005-09/2008.
• Statistical Models for Text-to-text Generation, Advanced Fellowship, funded by the EPSRC, 02/2005-01/2010.
• Application-based Text-to-Text Generation, Research Project Grant, funded by the EPSRC, 06/2006-06/2009.
• Robust Pragmatics for Narrative Text, funded by the EPSRC, 01/2002-03/2005.
I am a member of EdinburghNLP, one of the largest natural language processing groups in the world, and of the Institute for Language, Cognition and Computation (ILCC) at the School of Informatics, which brings together research on language, speech, cognition and machine learning.
The Generative AI
Laboratory (GAIL) at the University of Edinburgh
is a centre of excellence dedicated to researching all aspects of
generative artificial intelligence (AI) in society. Uniting the
diverse research expertise across the University with generative
AI at its core, GAIL taps into a thriving AI landscape with
recognised strengths in natural language processing, machine
learning, and data-driven innovation.
The Edinburgh Laboratory for Integrated Artificial Intelligence (ELIAI) at the School of Informatics is seeking to enhance neural network models with reasoning capabilities, a skill required to enable many AI applications.
Edinburgh is one of four partners in the Science of Fundamental AI Research (SOFAIR) Lab, a national AI research lab led by UCL with Cambridge and Oxford and funded by UKRI. The lab explores new model architectures and ways of training AI without vast centralised compute, and builds open models that run on widely available hardware.
The UKRI AI Centre for Doctoral Training (CDT) in Designing Responsible Natural Language Processing is a PhD training programme aiming to develop doctoral graduates that represent a new paradigm of interdisciplinary NLP researcher, who are ready to realise the full potential of NLP-based systems and enable richer interactions that allow genuine partnerships between humans and AI.
The Edinburgh Futures Institute (EFI), home to the Designing Responsible NLP CDT, brings together researchers from across the University to address complex societal challenges, with a focus on socially just AI and data.