Chenxin DIAO

I am a PhD student at the University of Edinburgh, supervised by Emily Allaway.

I am on a quest to understand when and why I can trust the words I read (whether written by humans or LLMs). This drives my work on modelling soundness: asking both whether a statement is correct and whether the interpretations are justified. Drawing on logic and linguistics, I develop neuro-symbolic methods to investigate these questions at scale. My current focus is over-generalisation in scientific writing, examining how claims become broader than the evidence supports.

I also maintain regular external research connections, including with Huawei Edinburgh Research Centre and CoCoA Lab led by my former supervisor, Steve Wilson. Before joining the University, I worked in industry on problems related to semantic parsing and agentic retrieval.

news

Sep 15, 2026	How broad is that claim? Mapping Generalisation in NLP Research by Chenxin Diao, Nataliya Stepanova, Emily Allaway will appear at EMNLP 2026. HarmReduction: Benchmarking LLMs in Harm Reduction Information Provision to Support People Who Use Drugs by Kaixuan Wang, Chenxin Diao, Jason T Jacques, Zhongliang Guo, Shuai Zhao will appear at EMNLP 2026.
Sep 01, 2025	I joined ILCC at the University of Edinburgh as a full-time PhD student, supervised by Emily Allaway.

selected publications

ACL Findings

GeAR: Graph-enhanced Agent for Retrieval-augmented Generation

Zhili Shen^†, Chenxin Diao^†, Pavlos Vougiouklis^†, and 8 more authors

In The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), 2025

DOI arXiv Bib Blog Poster

@inproceedings{shen2025gear,
  title = {GeAR: Graph-enhanced Agent for Retrieval-augmented Generation},
  author = {Shen, Zhili and Diao, Chenxin and Vougiouklis, Pavlos and Merita, Pascual and Piramanayagam, Shriram and Graux, Damien and Tu, Dandan and Jiang, Zeren and Lai, Ruofei and Ren, Yang and others},
  booktitle = {The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)},
  pages = {12049--12072},
  year = {2025},
  google_scholar_id_no_citation = {2osOgNQ5qMEC},
  doi = {10.18653/v1/2025.findings-acl.624},
}

EMNLP

Improving Retrieval-augmented Text-to-SQL with AST-based Ranking and Schema Pruning

Zhili Shen^†, Pavlos Vougiouklis^†, Chenxin Diao^†, and 3 more authors

In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024

DOI arXiv Bib Video Poster Slides

@inproceedings{shen2024improving,
  title = {Improving Retrieval-augmented Text-to-{SQL} with {AST}-based Ranking and Schema Pruning},
  author = {Shen, Zhili and Vougiouklis, Pavlos and Diao, Chenxin and Vyas, Kaustubh and Ji, Yuanyi and Pan, Jeff Z.},
  editor = {Al-Onaizan, Yaser and Bansal, Mohit and Chen, Yun-Nung},
  booktitle = {Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing},
  year = {2024},
  address = {Miami, Florida, USA},
  publisher = {Association for Computational Linguistics},
  url = {https://aclanthology.org/2024.emnlp-main.449/},
  doi = {10.18653/v1/2024.emnlp-main.449},
  pages = {7865--7879},
  google_scholar_id_no_citation = {d1gkVwhDpl0C},
}