CUHK MMLab · Meta Superintelligence Labs

Video foundation models, grounded multimodal agents, and real-world AI evaluation.

I am a final-year Ph.D. candidate at CUHK MMLab, advised by Prof. Dahua Lin. My research centers on vision-language models, multimodal agents, and long-horizon agent evaluation.

I am currently a Research Scientist Intern at Meta Superintelligence Labs, working on video grounding with Jie Lei. Previously, I worked on SAM3 with Nicolas Carion at Meta and on multimodal LLMs at Shanghai AI Laboratory with Xiaoyi Dong and Jiaqi Wang.

CV Scholar GitHub Email

Expected graduation: Summer 2027. Open to Research Scientist opportunities in multimodal AI, video VLMs, and agent evaluation.

2,211 Google Scholar citations
19 h-index
10 first / co-first author papers
ICLR · CVPR · ICCV plus NeurIPS, ICML, ECCV, ACL

Citation snapshot: June 9, 2026.

News

Latest updates

Released WildClawBench, a real-world long-horizon agent benchmark with 60 human-authored multimodal tasks.

SetCon is online, introducing set-level concept prediction for open-ended referring segmentation.

Three papers accepted at ICLR 2026: SAM3, SeC, and ScaleCap.

Released SAM3 for concept-level detection, segmentation, and tracking in images and videos.

Publications

Selected papers

* equal contribution · † project lead

SAM3 thumbnail

SAM3: Segment Anything with Concepts

Meta SAM3 Team, incl. Shuangrui Ding. ICLR 2026.

Project arXiv HF Code
WildClawBench thumbnail

WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation

Shuangrui Ding*†, Xuanlang Dai*, Long Xing*, et al. arXiv 2026.

Leaderboard arXiv Code
SAM2Long thumbnail

SAM2Long: Enhancing SAM 2 for Long Video Segmentation with a Training-Free Memory Tree

Shuangrui Ding, Rui Qian, Xiaoyi Dong, Pan Zhang, Yuhang Zang, Yuhang Cao, Yuwei Guo, Dahua Lin, Jiaqi Wang. ICCV 2025.

Project arXiv Code
SongComposer thumbnail

SongComposer: A Large Language Model for Lyric and Melody Generation in Song Composition

Shuangrui Ding*, Zihan Liu*, Xiaoyi Dong, Pan Zhang, Rui Qian, Junhao Huang, Conghui He, Dahua Lin, Jiaqi Wang. ACL 2025 Main.

arXiv Code Demo
Dispider thumbnail

Dispider: Enabling Video LLMs with Active Real-Time Interaction via Disentangled Perception, Decision, and Reaction

Rui Qian*, Shuangrui Ding*, Xiaoyi Dong, Pan Zhang, Yuhang Zang, Yuhang Cao, Dahua Lin, Jiaqi Wang. CVPR 2025.

arXiv Code
SeC thumbnail

SeC: Advancing Complex Video Object Segmentation via Progressive Concept Construction

Zhixiong Zhang*, Shuangrui Ding*, Xiaoyi Dong, Songxin He, Jianfan Lin, Junsong Tang, Yuhang Zang, Yuhang Cao, Dahua Lin, Jiaqi Wang. ICLR 2026.

Project arXiv Code

Background

Experience and education

Meta Superintelligence Labs, Bellevue

Research Scientist Intern on vision-language models for video grounding, supervised by Jie Lei.

The Chinese University of Hong Kong

Ph.D. in Information Engineering at MMLab, advised by Prof. Dahua Lin.

Shanghai AI Laboratory

Worked on LLMs and advanced video understanding with Xiaoyi Dong and Jiaqi Wang.

Shanghai Jiao Tong University

M.S. in Information and Communication Engineering. Graduate National Scholarship awardee.

University of Michigan

B.S.E. in Computer Science. Summa Cum Laude; GPA: 3.9/4.0.

Shanghai Jiao Tong University

B.S.E. in Electrical and Computer Engineering. National Scholarship Awardee; GPA: 3.8/4.0.

Honors

Selected awards

  • CUHK Vice-Chancellor's Ph.D. Scholarship, 80,000 HKD, 2023
  • Graduate National Scholarship, Top 2%, 2022
  • Shanghai Outstanding Graduate, Top 5%, 2021
  • Mathematical Contest in Modeling Finalist, Top 0.3%, 2019
  • Undergraduate National Scholarship, Top 2%, 2018

Talks and service

Community