Yucheng Du

yucheng-du

1

·

https://yucheng-du.github.io/

AI & ML interests

LLM reliability, mechanistic interpretability, AI auditing, computational linguistics

Recent Activity

submitted a paper 4 days ago

Recognition-Refusal Misalignment in LLMs: Why Models Answer Structurally Unanswerable Questions

authored a paper 4 days ago

Geometric Deviation as an Unsupervised Pre-Generation Reliability Signal: Probing LLM Representations for Answerability

authored a paper 4 days ago

Recognition-Refusal Misalignment in LLMs: Why Models Answer Structurally Unanswerable Questions

View all activity

Organizations

submitted a paper to Daily Papers 4 days ago

Recognition-Refusal Misalignment in LLMs: Why Models Answer Structurally Unanswerable Questions

Paper • 2608.29109 • Published 15 days ago • 17

authored 2 papers 4 days ago

Geometric Deviation as an Unsupervised Pre-Generation Reliability Signal: Probing LLM Representations for Answerability

Paper • 2605.03196 • Published May 4

Recognition-Refusal Misalignment in LLMs: Why Models Answer Structurally Unanswerable Questions

Paper • 2608.29109 • Published 15 days ago • 17