The Natural Language Processing (NLP) Group at KCL is comprised of PhD and postdoctoral students, professors and others who are interested in solving computational problems related to the understanding of human language. This encompasses a wide range of topics including sentiment analysis, topic/event extraction, question answering, cross-modal retrieval, text illustration, social media analysis and many more, typically approached with machine learning.
All images have been generated using DALL-E.
Projects

Event-Centric Framework for Natural Language Understanding
The five-year UKRI-funded Turing AI Fellowship awarded to Yulan He aims to develop a machine reading comprehension model in which a computer could continuously build and update a graph of eventualities as reading progresses.

New Language Modelling
Lin Gui and Yulan He have been awarded a prestigious EPSRC New Horizons grant for a high-risk research project with potentially transformative impact. The project aims to develop a new language modelling method allowing for a more faithful and explainable approximation for the input text.

Automated Scoring System for GCSE Science Exams
Funded by AQA, the project aims to develop an automated scoring system for assessing students’ answers to descriptive questions in GCSE Biology or Chemistry. The system is expected to produce prediction of marks and generate the rationales explaining the model decisions.

Character-Centric Narrative Understanding
The EPSRC ICASE project, jointly funded by Huawei London Research Centre, aims to develop new AI algorithms for automatic understanding of narratives in novels.

Model Interpretability
In our EPSRC-funded project, “Twenty20Insight”, we aim to investigate explainable AI (XAI) approaches which can provide interpretations both faithful to model decisions and are also better understood by humans.

PANACEA: PANdemic Ai Claim vEracity Assessment
Led by Yulan He, the EPSRC-funded PANACEA project developed novel supervised/unsupervised methods for veracity assessment of claims unverified at the time of posting, by integrating information from multiple sources and building a knowledge network that enables cross verification.
Activities

Tutorial on Uncertainty Quantification for Text Classification in SIGIR 2023
Yulan He and Lin Gui from King's College London, together with Dell Zhang, Murat Sensoy, Masoud Makrehchi, Bilyana Taneva-Popova at Thomson Reuters, will deliver a tutorial on Uncertainty Quantification for Text Classification in SIGIR 2023 in July 2023.

Invited Talk on Uncertainty Interpretation and Calibration of Language Models
Yulan He is invited to give a talk on uncertainty interpretation and calibration of language models in NLDB 2023 in June 2023.

Nine Papers Accepted to EACL 2023
The NLP group has nine papers accepted to EACL 2023.
- EACL 2023
- Event Temporal Relation Extraction with Bayesian Translational Model
- A User-Centered, Interactive, Human-in-the-Loop Topic Modelling System
- Distinguishability Calibration to In-Context Learning
- NapSS: Paragraph-level Medical Text Simplification via Narrative Prompting and Sentence-matching Summarization
- PANACEA: An Automated Misinformation Detection System on COVID-19
- K-hop neighbourhood regularization for few-shot learning on graphs: A case study of text classification
- CK-Transformer: Commonsense Knowledge Enhanced Transformers for Referring Expression Comprehension
- An Extended Sequence Tagging Vocabulary for Grammatical Error Correction
- Reading and Reasoning over Chart Images for Evidence-based Automated Fact-Checking

Invited Talk on NLP Research to Drive FinTech
Yulan He gave an invited talk on “NLP Research to Drive FinTech: Now and Next” in the Gillmore Centre for Financial Technology at Warwick Business School in December 2022.
Events

Long Narrative Understanding in the Era of Large Language Models
Prof Yulan He will present her group's latest research on Large Language Models
Please note: this event has passed.
Resources
Code & Dataset
- NapSS: Paragraph-level Medical Text Simplification via Narrative Prompting and Sentence-matching Summarization, Findings of EACL, May. 2023.
- Tracking Brand-Associated Polarity-Bearing Topics in User Reviews, Transactions of the Association for Computational Linguistics, accepted.
- PHEE: A Dataset for Pharmacovigilance Event Extraction from Text, The 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP), Dec. 2022.
- Event-Centric Question Answering via Contrastive Learning and Invertible Event Transformation, Findings of EMNLP, Dec. 2022.
- Hierarchical Interpretation of Neural Text Classification, Computational Linguistics, to appear.
- Cross-modal Prototype Driven Network for Radiology Report Generation, 17th European Conference on Computer Vision (ECCV), Oct. 2022.
- Addressing Token Uniformity in Transformers via Singular Value Transformation, 38th Conference on Uncertainty in Artificial Intelligence (UAI), Aug. 2022.
- Natural Language Inference with Self-Attention for Veracity Assessment of Pandemic Claims, 2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), Jul. 2022.
- Disentangled Learning of Stance and Aspect Topics for Vaccine Attitude Detection in Social Media, 2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), Jul. 2022.
- Extracting Event Temporal Relations via Hyperbolic Geometry, Conference on Empirical Methods in Natural Language Processing (EMNLP), Nov. 2021.
- Position Bias Mitigation: A Knowledge-Aware Graph Model for Emotion Cause Extraction, The 59th Annual Meeting of the Association for Computational Linguistics (ACL), Aug. 2021.
- A Neural Generative Model for Joint Learning Topics and Topic-Specific Word Embeddings, Transactions of the Association for Computational Linguistics, accepted.
Projects

Event-Centric Framework for Natural Language Understanding
The five-year UKRI-funded Turing AI Fellowship awarded to Yulan He aims to develop a machine reading comprehension model in which a computer could continuously build and update a graph of eventualities as reading progresses.

New Language Modelling
Lin Gui and Yulan He have been awarded a prestigious EPSRC New Horizons grant for a high-risk research project with potentially transformative impact. The project aims to develop a new language modelling method allowing for a more faithful and explainable approximation for the input text.

Automated Scoring System for GCSE Science Exams
Funded by AQA, the project aims to develop an automated scoring system for assessing students’ answers to descriptive questions in GCSE Biology or Chemistry. The system is expected to produce prediction of marks and generate the rationales explaining the model decisions.

Character-Centric Narrative Understanding
The EPSRC ICASE project, jointly funded by Huawei London Research Centre, aims to develop new AI algorithms for automatic understanding of narratives in novels.

Model Interpretability
In our EPSRC-funded project, “Twenty20Insight”, we aim to investigate explainable AI (XAI) approaches which can provide interpretations both faithful to model decisions and are also better understood by humans.

PANACEA: PANdemic Ai Claim vEracity Assessment
Led by Yulan He, the EPSRC-funded PANACEA project developed novel supervised/unsupervised methods for veracity assessment of claims unverified at the time of posting, by integrating information from multiple sources and building a knowledge network that enables cross verification.
Activities

Tutorial on Uncertainty Quantification for Text Classification in SIGIR 2023
Yulan He and Lin Gui from King's College London, together with Dell Zhang, Murat Sensoy, Masoud Makrehchi, Bilyana Taneva-Popova at Thomson Reuters, will deliver a tutorial on Uncertainty Quantification for Text Classification in SIGIR 2023 in July 2023.

Invited Talk on Uncertainty Interpretation and Calibration of Language Models
Yulan He is invited to give a talk on uncertainty interpretation and calibration of language models in NLDB 2023 in June 2023.

Nine Papers Accepted to EACL 2023
The NLP group has nine papers accepted to EACL 2023.
- EACL 2023
- Event Temporal Relation Extraction with Bayesian Translational Model
- A User-Centered, Interactive, Human-in-the-Loop Topic Modelling System
- Distinguishability Calibration to In-Context Learning
- NapSS: Paragraph-level Medical Text Simplification via Narrative Prompting and Sentence-matching Summarization
- PANACEA: An Automated Misinformation Detection System on COVID-19
- K-hop neighbourhood regularization for few-shot learning on graphs: A case study of text classification
- CK-Transformer: Commonsense Knowledge Enhanced Transformers for Referring Expression Comprehension
- An Extended Sequence Tagging Vocabulary for Grammatical Error Correction
- Reading and Reasoning over Chart Images for Evidence-based Automated Fact-Checking

Invited Talk on NLP Research to Drive FinTech
Yulan He gave an invited talk on “NLP Research to Drive FinTech: Now and Next” in the Gillmore Centre for Financial Technology at Warwick Business School in December 2022.
Events

Long Narrative Understanding in the Era of Large Language Models
Prof Yulan He will present her group's latest research on Large Language Models
Please note: this event has passed.
Resources
Code & Dataset
- NapSS: Paragraph-level Medical Text Simplification via Narrative Prompting and Sentence-matching Summarization, Findings of EACL, May. 2023.
- Tracking Brand-Associated Polarity-Bearing Topics in User Reviews, Transactions of the Association for Computational Linguistics, accepted.
- PHEE: A Dataset for Pharmacovigilance Event Extraction from Text, The 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP), Dec. 2022.
- Event-Centric Question Answering via Contrastive Learning and Invertible Event Transformation, Findings of EMNLP, Dec. 2022.
- Hierarchical Interpretation of Neural Text Classification, Computational Linguistics, to appear.
- Cross-modal Prototype Driven Network for Radiology Report Generation, 17th European Conference on Computer Vision (ECCV), Oct. 2022.
- Addressing Token Uniformity in Transformers via Singular Value Transformation, 38th Conference on Uncertainty in Artificial Intelligence (UAI), Aug. 2022.
- Natural Language Inference with Self-Attention for Veracity Assessment of Pandemic Claims, 2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), Jul. 2022.
- Disentangled Learning of Stance and Aspect Topics for Vaccine Attitude Detection in Social Media, 2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), Jul. 2022.
- Extracting Event Temporal Relations via Hyperbolic Geometry, Conference on Empirical Methods in Natural Language Processing (EMNLP), Nov. 2021.
- Position Bias Mitigation: A Knowledge-Aware Graph Model for Emotion Cause Extraction, The 59th Annual Meeting of the Association for Computational Linguistics (ACL), Aug. 2021.
- A Neural Generative Model for Joint Learning Topics and Topic-Specific Word Embeddings, Transactions of the Association for Computational Linguistics, accepted.