Ruixuan Yang

Undergraduate Student

About Me

I am Ruixuan Yang, an undergraduate in the School of Mathematics and Statistics at Xi’an Jiaotong University (XJTU). I’m interested in Model Efficiency, Multimodal LLMs (MLLMs), Post-training & Reasoning, etc.

I am currently a visiting student in Prof. Yulun Zhang’s group at Shanghai Jiao Tong University and Prof. Huan Wang’s ENCODE Lab at Westlake University.

Education

B.S. in progress

Xi'an Jiaotong University

School of Mathematics and Statistics

Research Interests

MLLMs Model Efficiency Post-training & Reasoning Measuring AI Intelligence

Research

MLLMs
Model Efficiency
Post-training & Reasoning
Measuring AI Intelligence

I am also open to related questions across machine learning and AI.

Selected Publications

LT-OPD framework with a low-token student, a frozen full-token teacher, and a token-budget curriculum

arXiv 2026 Preprint

Fewer Tokens, More Self-Teaching: On-Policy Self-Distillation for Extreme Visual Token Reduction

Junxian Li^*, Ruixuan Yang^*, Tianao Zhang, Tiange Xu, Weisheng Dong, Yulun Zhang

* Equal contribution

Recovering multimodal capabilities under extreme visual-token reduction through on-policy self-distillation and a progressive token-budget curriculum.

PDF arXiv Code

ECHO architecture integrating a hierarchical memory system with a vision-language-action model

NeurIPS 2026 Accepted

ECHO: Continuous Hierarchical Memory for Vision-Language-Action Models

Yanbin Hu^*, Jin Cui^*, Jiayi Lu, Ruixuan Yang, Jun Ye, Boran Zhao, Xingyu Chen, Xuguang Lan, Pengju Ren

* Equal contribution

Organizing robot experiences in a continuous hierarchical memory to support retrieval and generalization in long-horizon tasks.

PDF arXiv

Overview of the COMPACT multi-teacher reasoning distillation framework

IJCAI 2026 Accepted

"The Whole Is Greater Than the Sum of Its Parts": A Compatibility-Aware Multi-Teacher CoT Distillation Framework

Jin Cui^*, Jiaqi Guo^*, Ruixuan Yang, Jiepeng Zhou, Jiayi Lu, Jiajun Xu, Jiangcheng Song, Boran Zhao, Pengju Ren

* Equal contribution

Combining complementary teachers through student-aware supervision for more effective reasoning distillation.

PDF arXiv

Overview of the MIND capability-aware multi-perspective distillation framework

ACL 2026 Main Conference · Long Paper

MIND: From Passive Mimicry to Active Reasoning through Capability-Aware Multi-Perspective CoT Distillation

Jin Cui^*, Jiaqi Guo^*, Jiepeng Zhou, Ruixuan Yang, Jiayi Lu, Jiajun Xu, Jiangcheng Song, Boran Zhao, Pengju Ren

* Equal contribution

Aligning diverse reasoning perspectives with the student's evolving capabilities to move beyond passive imitation.

PDF ACL Anthology arXiv

News

Date	Update
Sep 2026	Our preprint LT-OPD, Fewer Tokens, More Self-Teaching, is now available on arXiv.
Sep 2026	ECHO was accepted to NeurIPS 2026.
Apr 2026	COMPACT was accepted to IJCAI 2026.
Apr 2026	MIND was accepted to ACL 2026 as a main-conference long paper.

Experience

Visiting Student

Shanghai Jiao Tong University

Jul 2026 – Present

Prof. Yulun Zhang’s group. Advised by Prof. Yulun Zhang.

Visiting Student

Westlake University

Apr 2026 – Present

ENCODE Lab. Advised by Prof. Huan Wang.

Visiting Student

Xi'an Jiaotong University

Oct 2025 – May 2026

Cognitive Architecture Group, Institute of AI & Robotics. Advised by Prof. Pengju Ren and Boran Zhao.

Hobbies

Outside research, I enjoy basketball, traveling, and music, especially rock, hip-hop, and jazz. My favorite singers are Ringo Sheena (椎名林檎) and Deserts Chang (张悬).