About me

I am a Computer Science undergraduate at Peking University, expected to graduate in 2028.

My current research interests center on recursive self-improvement (RSI). I am exploring this direction with an open focus across agents and models.

Previously, I worked on multimodal models and embodied intelligence, particularly vision-language-action systems and navigation. My experience spans model post-training, data construction, evaluation, and AI agent development.

Papers

AAAI 2027Under review

ASCRIBE: Retrospective Responsibility Routing Post-Training for Hierarchical VLA System

Yuxing Long^*, Junyang Wang^*, Bin Cheng, Lei Kang, Xionghao Wu, Ziyan Yu, Enjin Zhou, Junhan Wang, Jiyao Zhang, Hao Dong^†

NeurIPS 2026

One Scan is Enough: Demonstration-Free Adaptation for Language-Guided Navigation

Zhuoyuan Yu^*, Yuxing Long^*, Du Wang^*, Junyang Wang, Hongwei Fan, Hao Dong^†

ICLR 2027Under review

DualFuture: Event-Triggered Dual-Future Residual Correction for Vision-Language-Action Policy

Yuxing Long^*, Ziyan Yu^*, Bin Cheng^*, Junyang Wang, Lei Kang, Hao Dong^†

AAAI 2027Under review

VERDICT: Unleashing Cross-Method Complementarity for Spatio-Temporal Video Grounding through Debate

Tao Shen, Minghang Zheng, Junyang Wang, Jingli Wei, Hong-Yi Yang, Yang Liu^†

* Equal contribution. † Corresponding author.

Research & industry experience

Full CV

Research

Dec 2025 — Sep 2026

Center on Frontiers of Computing Studies, Peking University

Undergraduate Research Assistant

Advisor: Hao Dong

Research on post-training for vision-language-action systems and language-guided navigation.

Mar 2026 — Jun 2026

Wangxuan Institute of Computer Technology, Peking University

Research Rotation

Advisor: Yang Liu

Research on spatio-temporal video grounding through multi-agent debate.

Industry

Jun 2026 — Sep 2026

JD Technology

Embodied AI Algorithm Engineering Intern

Worked on long-horizon task data, chain-of-thought synthesis, distributed SFT/RL post-training with Qwen-VL models, evaluation, and failure analysis.

Nov 2025 — Mar 2026

Ubiquant Innovation

AI Agent Engineering Intern

Contributed to Framía and One, building capabilities for content understanding, page generation, visual layout, research, and complex office tasks.

Selected projects

All projects

Reinforcement learning

CritPT-RL

An end-to-end GRPO post-training and evaluation pipeline for scientific coding tasks, with synthetic data, reward functions, and executable verifiers.

Project details Code

Research tools

README-AI

An AI paper reader for translating, discussing, and annotating research papers, with contextual figure explanations and related-work search.

Project details

Community

I help organize PLib, an AI student community at Peking University. Through P-Talk, we bring students together for talks on large language models and AI agents.

I also contribute to P-Lib, a course notes and resources platform with more than 5,000 registered users.

Core organizer · Aug 2025 — Present