Qiguang Chen

Qiguang Chen陈麒光

PhD Candidate in Computer Science

Harbin Institute of Technology · SCIR · Harbin, China · charleschen2333@gmail.com

I work on multi-modal, multi-lingual and chain-of-thought reasoning for large language models — building methods and benchmarks that make reasoning measurable, controllable and evaluable across languages and modalities. I am a PhD student at SCIR, Harbin Institute of Technology, advised by Prof. Wanxiang Che and Prof. Libo Qin.

Previously I did research internships at ByteDance Seed, Shanghai AI Laboratory and Baidu NLP. I also serve as Deputy Secretary-General of the MLNLP community. If you are seeking any form of academic cooperation, please feel free to email me.

Currently exploring RSI (reinforcement learning with self-improvement) · AI for Science · AI for Engineering — agents that improve themselves and accelerate scientific discovery. Happy to discuss!
Self-Improving Agents (RSI) AI for Science AI for Engineering Multi-modal / Multilingual Reasoning Evaluation & Benchmarks
Baidu Scholarship Nominee (Global Top 20) 2026 · 第十三届百度奖学金 · top-20 nominee among worldwide applicants
18
First / Co-first Author
9
Oral Papers
3
Best Paper / Poster Awards
Top-20
Baidu Scholarship 2026

🔥News

  • 2026.11 🎉 I am named a Top-20 nominee of the 2026 Baidu Scholarship.
  • 2026.08 🎉 We release VibeLifeBench, a benchmark for proactive and persistent life agents in a living world.
  • 2026.08 🎉 Our OffSeeker is accepted by EMNLP 2026, and Learning the Boundary of Solvability is accepted by EMNLP 2026 (Findings).
  • 2026.07 🎉 We release RSIBench-Data for benchmarking data-centric research in recursive self-improvement.
  • 2026.05 🎉 Our RBF++ is accepted by TPAMI 2026 (IF 18+). Our AI4Research Survey is accepted by SCIS 2026.
  • 2026.04 🎉 Our OMIBench and Dashboard2Code are accepted by ACL 2026 (Oral). Our UL-XCoT is accepted by ACL 2026.
  • 2026.02 🎉 Congratulations to our collaborators! Our Think-with-Video is accepted by CVPR 2026.
  • 2026.01 🎉 Congratulations to our collaborators! Our Refusal Index is accepted by ICLR 2026.

🚀Selected Work

NeurIPS 2024 First Author Oral · Top 1.8% accepted

Unlocking the Capabilities of Thought: A Reasoning Boundary Framework to Quantify and Optimize Chain-of-Thought

Qiguang Chen, Libo Qin, Jiaqi Wang, Jinxuan Zhou, Wanxiang Che

A reasoning boundary framework that quantifies how far a model's chain-of-thought can go and optimizes it — turning an open-ended ability into a measurable quantity.

Reasoning Boundary Framework: overview of the proposed framework, its combination law and the resulting categories of reasoning boundaries.
SCIS 2026 First Author LMG Best Poster

Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models

Qiguang Chen, Libo Qin, Jinhao Liu, Dengyun Peng, Jiannan Guan, Peng Wang, Mengkang Hu, Yuhang Zhou, Te Gao, Wanxiang Che

A systematic survey of long chain-of-thought reasoning — its landscape, training recipes and evaluation — with an accompanying paper list that has become a widely used reference for the community.

The long chain-of-thought survey teaser: deep reasoning, feasible reflection and extensive exploration as three frontiers of long-CoT research.
SCIS 2026 First Author 950+ papers surveyed

AI4Research: A Survey of Artificial Intelligence for Scientific Research

Qiguang Chen*, Mingda Yang*, Libo Qin, Jinhao Liu, Zheng Yan, Jiannan Guan, Dengyun Peng, Yiyan Ji, Hanjing Li, Mengkang Hu, Yimeng Zhang, Yihao Liang, Yuhang Zhou, Jiaqi Wang, Zhi Chen, Wanxiang Che

A data-centric review of AI for scientific research — organised around tasks, data and evaluation — released together with a living paper collection of 950+ papers.

The AI4Research taxonomy: AI for scientific comprehension, academic survey, discovery, writing and peer review.
ACL 2024 First Author Oral · Top 2.3% submitted

M3CoT: A Novel Benchmark for Multi-Domain Multi-step Multi-modal Chain-of-Thought

Qiguang Chen, Libo Qin, Jin Zhang, Zhi Chen, Xiao Xu, Wanxiang Che

The first benchmark that pushes multi-modal chain-of-thought to multi-domain and multi-step visual reasoning — revealing that VLLMs still answer from textual shortcuts and leave a wide gap to human reasoning.

M3CoT teaser: absence of visual-modal reasoning, single-step visual reasoning and multi-step visual reasoning compared.

📚Publications

Classified by main research concern (Partial)

18 first / co-first 9 orals 3 best paper / poster awards 7 research areas

🎯 Chain-of-Thought & Reasoning

🕹️ Agents

AI4Research

👀 Multi-Modal Modelling & Reasoning

🏳️‍🌈 Multilingual Modelling & Reasoning

🎙️ Dialogue System

Others

⭐Open-Source Projects

  • MLNLP SimBiber — a tool for simplifying BibTeX entries with official venue information.
  • MLNLP Top-AI-Conferences-Paper-with-Code — a curated list of AI conference papers with released code.
  • MLNLP AI-Paper-Collector — fully automated scripts for collecting AI-related papers.
  • SCIR HuoZi — HuoZi, a general-purpose Chinese large language model.
  • Baidu Trust-AI — PaddlePaddle’s trustworthy-AI toolkit.

🏆Honors & Awards

Baidu Scholarship Nominee (Global Top 20)
第十三届百度奖学金 · top-20 nominee among worldwide applicants
2026
Young Elite Scientists Sponsorship Program by CAST
Doctoral Program, China Association for Science and Technology
2025
National Scholarship
National-level scholarship for outstanding graduate students
2024
Paper Digest Most Influential AI Papers
Top-6, Most Influential arXiv AI Papers (Sep 2025) — for the long chain-of-thought survey
2025
  • 2023Graduate Special Scholarship — Harbin Institute of Technology
  • 2023Outstanding Graduates — Harbin Institute of Technology
  • 2022CCF Excellent College Students Prize — only 100 CS students in China were invited
  • 2022EMNLP MMNLU Workshop Best Paper
  • 20221st place in MMNLU Workshop Full-shot Setting
  • 2021National Second Prize — Contemporary Undergraduate Mathematical Contest in Modeling, China

🎓Education & Experience

2023.09 – Now
PhD Student, Computer Science
Harbin Institute of Technology · SCIR, advised by Prof. Wanxiang Che and Prof. Libo Qin
2019.09 – 2023.06
B.Eng., Computer Science
Harbin Institute of Technology

Work Experience

  • 2025.07 – 2026.04ByteDance Seed — research intern, Eval / Horizon team, Beijing
  • 2024.07 – 2024.10Shanghai AI Laboratory — research intern, Shanghai
  • 2022.06 – 2022.12Baidu NLP — research intern, Beijing

🛠️Academic Service

NLP Reviewer
ACL ARR 2024–2025 EMNLP ARR 2024–2025 NAACL ARR 2024–2025 NLPCC 2024–2025 IEEE/ACM TASLP
ML Reviewer
NeurIPS 2024–2025 ICLR 2025–2026 ICML 2025
AI Reviewer
Nature Machine Intelligence AAAI 2026 ACM MM 2025 Information Sciences Neurocomputing PeerJ Computer Science
Community
MLNLP 2022/2023 Outstanding Organizer Deputy Secretary-General, MLNLP

🤝Undergraduate Collaborators

I enjoy working with passionate undergraduates who have strong coding and math skills and bring fresh inspiration. If you are interested in collaborating, please feel free to email me.

  • Wenbo PanHIT → CityUHK (PhD)
  • Yuhang ZhouHIT → FDU (Master)
  • Jin Zhang & Zhixiong ZhangHIT → Shanghai AI Lab (PhD)
  • Yongheng ZhangAAU → CSU (Master)
  • Peng WangAAU → MUST (Master)
  • Jinhao LiuHIT → HIT (PhD)
  • Dengyun PengHIT (Weihai) → HIT (Master)
  • Yiyan JiHIT → NJU (PhD)
  • Yizhuo LiHIT → SJTU (PhD)
  • Chenyuan ZhangHIT → HIT Shenzhen (PhD)
  • Qionglin QiuCSU → HIT Shenzhen (PhD)

🌎Visitors