activity
20242026
collaborators

9 papers

cs.AI2026

AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent

Weidi Luo, Qiming Zhang, Yihao Quan +5

Coding agents based on large language models (LLMs) demonstrate remarkable autonomous capabilities, but they also introduce significant safety and misuse risks during multi-turn in…

cs.CV2026

MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory

Minghao Guo, Qingyue Jiao, Zeru Shi +14

Long-term agent memory is increasingly multimodal, yet existing evaluations rarely test whether agents preserve the visual evidence needed for later reasoning. In prior work, many…

cs.CL2026

Where Fake Citations Are Made: Tracing Field-Level Hallucination to Specific Neurons in LLMs

Yuefei Chen, Yihao Quan, Xiaodong Lin +1

LLMs frequently generate fictitious yet convincing citations, often expressing high confidence even when the underlying reference is wrong. We study this failure across 9 models an…

cs.CL2026

ART: Attention Replacement Technique to Improve Factuality in LLMs

Ziqin Luo, Yihao Quan, Xiaofeng Zhang +2

Hallucination in large language models (LLMs) continues to be a significant issue, particularly in tasks like question answering, where models often generate plausible yet incorrec…

cs.CV2026

Reinforcing Consistency in Video MLLMs with Structured Rewards

Yihao Quan, Zeru Shi, Jinman Zhao +1

Multimodal large language models (MLLMs) have achieved remarkable progress in video understanding. However, seemingly plausible outputs often suffer from poor visual and temporal g…

cs.CV2026

Counting Circuits: Mechanistic Interpretability of Visual Reasoning in Large Vision-Language Models

Liwei Che, Zhiyu Xue, Yihao Quan +7

Counting serves as a simple but powerful test of a Large Vision-Language Model's (LVLM's) reasoning; it forces the model to identify each individual object and then add them all up…