publications

peer-reviewed

1. UAI

   

   Semantic Self-Distillation for Language Model Uncertainty

   Edward Phillips, Sean Wu, Fredrik K. Gustafsson, and 2 more authors

   Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, 17–21 Aug 2026

   Abs arXiv Bib PMLR PDF Poster

   Large language models present challenges for principled uncertainty quantification, in part due to their complexity and the diversity of their outputs. Semantic dispersion, or the variance in the meaning of sampled answers, has been proposed as a useful proxy for model uncertainty, but the associated computational cost prohibits its use in latency-critical applications. We show that sampled semantic distributions can be distilled into lightweight student models which estimate a prompt-conditioned density before the language model generates an answer token. The student model predicts a semantic distribution over possible answers; the entropy of this distribution provides a prompt-level uncertainty signal, and the probability density allows answer-level reliability evaluation. Across experiments on TriviaQA and MMLU, we find our student models perform competitively relative to the teacher’s sampled semantic dispersion on a hallucination prediction task, whilst offering additional uncertainty primitives for out-of-domain detection and multiple-choice answer selection. We term this technique Semantic Self-Distillation (SSD), which can serve as a general framework for distilling predictive uncertainty in complex output spaces beyond language.

   @inproceedings{phillips2026ssd,
     title = {Semantic Self-Distillation for Language Model Uncertainty},
     author = {Phillips, Edward and Wu, Sean and Gustafsson, Fredrik K. and Gao, Boyan and Clifton, David A.},
     booktitle = {Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence},
     editor = {Perković, Emilija and Malinsky, Daniel},
     volume = {337},
     series = {Proceedings of Machine Learning Research},
     pages = {5427--5447},
     month = {17--21 Aug},
     publisher = {PMLR},
     year = {2026},
   }

2. TMLR

   

   Geometric Uncertainty for Detecting and Correcting Hallucinations in LLMs

   Edward Phillips, Sean Wu, Soheila Molaei, and 3 more authors

   Transactions on Machine Learning Research, 2026

   arXiv Bib OpenReview

   @article{phillips2026geometric,
     title = {Geometric Uncertainty for Detecting and Correcting Hallucinations in {LLM}s},
     author = {Phillips, Edward and Wu, Sean and Molaei, Soheila and Belgrave, Danielle and Thakur, Anshul and Clifton, David A.},
     journal = {Transactions on Machine Learning Research},
     issn = {2835-8856},
     year = {2026},
   }

3. COLM

   

   BAS: A Decision-Theoretic Approach to Evaluating Large Language Model Confidence

   Sean Wu, Fredrik K. Gustafsson, Edward Phillips, and 3 more authors

   Conference on Language Modeling (COLM), 2026

   arXiv Bib OpenReview

   @inproceedings{wu2026bas,
     title = {BAS: A Decision-Theoretic Approach to Evaluating Large Language Model Confidence},
     author = {Wu, Sean and Gustafsson, Fredrik K. and Phillips, Edward and Gao, Boyan and Thakur, Anshul and Clifton, David A.},
     booktitle = {Conference on Language Modeling (COLM)},
     year = {2026},
   }

preprints & under review

1. arXiv

   When Text and Numbers Disagree: Evidence Arbitration in Large Language Models

   Mattia Carletti, Edward Phillips, Fredrik K. Gustafsson, and 5 more authors

   2026

   arXiv Bib

   @misc{carletti2026arbitration,
     title = {When Text and Numbers Disagree: Evidence Arbitration in Large Language Models},
     author = {Carletti, Mattia and Phillips, Edward and Gustafsson, Fredrik K. and Palo, Patitapaban and Clifton, Lei and Belgrave, Danielle and Gu, Xiao and Clifton, David A.},
     year = {2026},
   }

2. arXiv

   Entropy Alone is Insufficient for Safe Selective Prediction in LLMs

   Edward Phillips, Fredrik K. Gustafsson, Sean Wu, and 2 more authors

   2026

   arXiv Bib

   @misc{phillips2026entropy,
     title = {Entropy Alone is Insufficient for Safe Selective Prediction in LLMs},
     author = {Phillips, Edward and Gustafsson, Fredrik K. and Wu, Sean and Thakur, Anshul and Clifton, David A.},
     year = {2026},
   }

3. TechRxiv

   Foundation Models for Biosignals: A Survey

   Xiao Gu, Yuxuan Shu, Jinpei Han, and 16 more authors

   2026

   Bib TechRxiv

   @misc{gu2026biosignals,
     title = {Foundation Models for Biosignals: A Survey},
     author = {Gu, Xiao and Shu, Yuxuan and Han, Jinpei and Liu, Yuxuan and Liu, Zhangdaihong and Anibal, James and Sangha, Veer and Phillips, Edward and Segal, Bradley and Yuan, Hang and Liu, Fenglin and Branson, Kim and Schwab, Patrick and Belgrave, Danielle and Clifton, Lei and Spathis, Dimitris and Lampos, Vasileios and Faisal, A. Aldo and Clifton, David A.},
     year = {2026},
   }