Yihao Quan

I am a 1st Year Ph.D. student at Rutgers University, working on multimodal large language models with Prof. Ruixiang (Ryan) Tang.

I am currently a Research Scientist Intern with the AMD GenAI Team. Before Rutgers, I obtained my B.S. degree from Beijing Jiaotong University in 2025.

qyhhere[at]gmail[dot]com | Google Scholar | GitHub | LinkedIn | X | WeChat

news

Jul 2026

From Plausible to Grounded was accepted to COLM 2026.

Jun 2026

Counting Circuits was accepted to ECCV 2026.

May 2026

Started as a Research Scientist Intern at Advanced Micro Devices, AMD GenAI Team.

Apr 2026

ART was accepted to ACL 2026 Main.

Aug 2025

Seeing Clearly by Layer Two was accepted to EMNLP 2025 Oral.

Jan 2025

From Redundancy to Relevance was accepted to NAACL 2025 Oral.

selected papers

* = equal contribution.

Preprint

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

Weidi Luo, Qiming Zhang, Yihao Quan, Mingyu Jin, Jie Cai, Chaowei Xiao, Jingcheng Niu, Zhen Xiang

paper code
Preprint

Improving Visual Reasoning with Iterative Evidence Refinement

Zeru Shi*, Kai Mei*, Yihao Quan, Dimitris N. Metaxas, Ruixiang Tang

paper
Preprint

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

Minghao Guo*, Qingyue Jiao*, Zeru Shi*, Yihao Quan, Boxuan Zhang, Danrui Li, Liwei Che, Wujiang Xu, Shilong Liu, Zirui Liu, Mubbasir Kapadia, Vladimir Pavlovic, Jiang Liu, Mengdi Wang, Yiyu Shi, Dimitris N. Metaxas, Ruixiang Tang

paper code
COLM 2026

From Plausible to Grounded: Reinforcing Consistency in Video MLLMs with Structured Rewards

Yihao Quan, Zeru Shi, Jinman Zhao, Ruixiang Tang

paper
EMNLP 2025 Oral

Seeing Clearly by Layer Two: Enhancing Attention Heads to Alleviate Hallucination in LVLMs

Yihao Quan*, Xiaofeng Zhang*, Chaochen Gu, Chen Shen, Xiaosong Yuan, Shaotian Yan, Jieping Ye

paper code
ACL 2026 Main

ART: Information Flow in LVLMs Across Reasoning Tasks

Ziqin Luo*, Yihao Quan*, Xiaofeng Zhang*, Xiaosong Yuan, Chen Shen

paper code
NAACL 2025 Oral

From Redundancy to Relevance: Information Flow in LVLMs Across Reasoning Tasks

Xiaofeng Zhang*, Yihao Quan*, Chaochen Gu, Chen Shen, Xiaosong Yuan, Shaotian Yan, Liang Xie, Wenxiao Wang, Hao Tang, Jieping Ye

paper code
ECCV 2026

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

Liwei Che*, Zhiyu Xue*, Yihao Quan*, Benlin Liu, Zeru Shi, Ruixiang Tang, Ranjay Krishna, Vladimir Pavlovic

paper
AAAI 2025

Enhancing Multimodal Large Language Models Complex Reason via Similarity Computation

Xiaofeng Zhang*, Fanshuo Zeng*, Yihao Quan, Zheng Hui, Jiawei Yao

paper code

experience

AMD

Research Sci. Intern

2026.06 - Present

KAUST

Research Asst.

2024.09 - 11

Alibaba

Research Intern

2024.06 - 09

SJTU

Research Asst.

2024.02 - 06

photos