Publications

2026

1. EMNLP 2026

   

   Understanding Why Language Models Hallucinate: Testing Reasoning Against Priors

   Yangfan Hu^*, Xuhan Tong^*, Haoyue Bai^*, and 5 more authors

   2026

   Accepted to Findings of EMNLP 2026

   arXiv Bib HTML PDF

   @misc{hu2026understandingwhylanguagemodelshallucinate,
     title = {Understanding Why Language Models Hallucinate: Testing Reasoning Against Priors},
     author = {Hu, Yangfan and Tong, Xuhan and Bai, Haoyue and Ding, Xi and Bharadwaj, Shashank Muralidhar and Cao, Siyang and Nowak, Robert and Zhang, Jiawei},
     year = {2026},
     note = {Accepted to Findings of EMNLP 2026},
     archiveprefix = {arXiv},
     primaryclass = {cs.CL},
     url = {https://arxiv.org/abs/2607.00447},
   }

2. CCS 2026

   

   Understanding and Preserving Safety in Fine-Tuned LLMs

   Jiawen Zhang^*, Yangfan Hu^*, Kejia Chen, and 7 more authors

   2026

   Accepted to ACM CCS 2026

   arXiv Bib PDF

   @misc{zhang2026understandingpreservingsafetyfinetuned,
     title = {Understanding and Preserving Safety in Fine-Tuned {LLMs}},
     author = {Zhang, Jiawen and Hu, Yangfan and Chen, Kejia and He, Lipeng and Ma, Jiachen and Lou, Jian and Li, Dan and Liu, Jian and Yang, Xiaohu and Jia, Ruoxi},
     year = {2026},
     note = {Accepted to ACM CCS 2026},
     archiveprefix = {arXiv},
     primaryclass = {cs.LG},
     url = {https://arxiv.org/abs/2601.10141},
   }