| Jan 30, 2026 | Our new preprint Reasoning While Asking: Transforming Reasoning Large Language Models from Passive Solvers to Proactive Inquirers is released. We propose a Proactive Interactive Reasoning framework that enables reasoning-LLMs to proactively ask clarification questions to resolve user ambiguity, significantly improving accuracy and efficiency. |
| Jan 26, 2026 | Our work Neuron-Aware Data Selection in Instruction Tuning for Large Language Models has been accepted by ICLR 2026. We propose the Nait framework, which selects optimal instruction-tuning data samples by analyzing the similarity of neuron activation patterns. |
| Sep 01, 2025 | Our work RepreGuard: Detecting LLM-Generated Text by Revealing Hidden Representation Patterns will be presented at EMNLP 2025. Looking forward to insightful discussions at the conference — see you in Suzhou, China! |
| Aug 01, 2025 | One paper accepted by Transactions of the Association for Computational Linguistics (TACL) as first author: RepreGuard: Detecting LLM-Generated Text by Revealing Hidden Representation Patterns. We propose RepreGuard, a low-overhead, highly generalizable, and interpretable detector based on the observation that there are significant differences in neural activation patterns when LLMs process LLM-generated text versus human-written text. |
| Apr 03, 2025 | Our new preprint Understanding Aha Moments: From External Observations to Internal Mechanisms is released. We reveal that “aha moments” help LRMs solve complex problems by using anthropomorphic language, adjusting uncertainty, and preventing reasoning collapse, while internally balancing self-reflection with reasoning and adapting to task difficulty. |
| Jun 15, 2024 | Our paper related to Explainable AI (XAI) is out: A Hopfieldian View-based Interpretation for Chain-of-Thought Reasoning. |