Bingshuai Liu

Bingshuai Liu

Master Student, Xiamen University

About me

I am a third-year master’s degree student of Xiamen University, advised by Dr. Jinsong Su. I received my bachelor degree in School of Informatics, Xiamen University.

News

  • [2026.01] Our DHPO is released! Orchestrating Tokens and Sequences: Dynamic Hybrid Policy Optimization for RLVR. arXiv

  • [2025.09] 🎉 Our new work SPEC-RL: Accelerating On-Policy Reinforcement Learning with Speculative Rollouts is released!
    SPEC-RL introduces speculative decoding into RLVR, achieving 2–3× rollout acceleration without loss in reasoning accuracy.
    🔗 GitHub Repository | Project Page

Publications & Preprints

Orchestrating Tokens and Sequences: Dynamic Hybrid Policy Optimization for RLVR. arXiv

Zijun Min, Bingshuai Liu, Ante Wang, Long Zhang, Anxiang Zeng, Haibo Zhang, Jinsong Su.

SPEC-RL: Accelerating On-Policy Reinforcement Learning via Speculative Rollouts. arXiv

Bingshuai Liu, Ante Wang, Zijun Min*, Liang Yao, Haibo Zhang, Yang Liu, Xu Han, Peng Li, Anxiang Zeng, Jinsong Su.

EditEval: Towards Comprehensive and Automatic Evaluation for Text-guided Video Editing. ACM MM 2025 Regular Paper

Bingshuai Liu*, Ante Wang, Zijun Min, Chenyang Lyu, Longyue Wang, Zhihao Wang, Xu Han, Peng Li, Jinsong Su.

On the cultural gap in text-to-image generation. ECAI 2024

Bingshuai Liu*, Longyue Wang*, Chenyang Lyu, Yong Zhang, Jinsong Su, Shuming Shi, and Zhaopeng Tu.

Exploring Optimal Transport-Based Multi-Grained Alignments for Text-Molecule Retrieval. BIBM 2024 Regular Paper

Zijun Min*, Bingshuai Liu*, Liang Zhang, Jia Song, Jinsong Su, Song He, and Xiaochen Bo.

Retrieval-augmented Multi-modal Chain-of-Thoughts Reasoning for Large Language Models. IJCNN 2025

Bingshuai Liu*, Chenyang Lyu*, Zijun Min, Zhanyu Wang, Jinsong Su, and Longyue Wang.

Macaw-LLM: Multi-Modal Language Modeling with Image, Audio, Video, and Text Integration.

Lyu, Chenyang, Minghao Wu, Longyue Wang, Xinting Huang, Bingshuai Liu, Zefeng Du, Shuming Shi, and Zhaopeng Tu.

Invited Talks

  • 2025-10: SPEC-RL: Accelerating On-Policy Reinforcement Learning via Speculative Rollouts, StepFun | [slide]

Internships

  • 2022.08-2023.07: Tencent AI Lab, Shenzhen. Mentors: Dr. Longyue Wang and Dr. Zhaopeng Tu.
  • 2024-07-2025-04: Li Auto, Beijing. Mentor: Hao Xu.
  • 2025-06-: Shopee LLM Team, Shanghai. Mentor: Haibo Zhang.