🧑‍🔬 Shixiang Tang

🏛️ Affiliations

🎓Chinese University of Hong KongHong Kong · China2024–2026 · 20 paperscurrent🔬Shanghai Artificial Intelligence LaboratoryShanghai · China2024–2026 · 16 paperscurrent🎓University of Hong KongPok Fu Lam · China2026 · 1 papercurrent🎓The University of SydneySydney · Australia2021–2024 · 17 papers🏢SensetimeBeijing · China2020–2023 · 7 papers🎓Fudan UniversityShanghai · China2012–2021 · 2 papersEducation 2012–2016 · Bachelor in Science, Physics🔬State Key Laboratory of Surface PhysicsShanghai2020–2021 · 2 papers

🏛️The University of Sydney - The QuarterSydney · Australia2020–2023Education 2020–2023 · Ph.D candidate, School of Electrical and Information Engineering

🏛️The Chinese University of Hong Kong Graduate SchoolHong Kong · China2016–2018Education 2016–2018 · Master of Philosophy in Physics, Physics

🎓University of OxfordOxford · United Kingdom2018Employment since 2018 · Research Intern, Smart City Group

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine.

🧭 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📅 Works by year

2020s

📄 Papers

The most cited first (OpenAlex and our library).

1. Many Heads Are Better Than One: Improved Scientific Idea Generation by A LLM-Based Multi-Agent System (2025 · 40 citations)
   Most LLM tools for automated research ideation use a single agent or rely on artificial collaboration setups, such as invented researcher profiles or synthetic co-author networks.
2. ResearchBench: Benchmarking LLMs in Scientific Discovery via Inspiration-Based Task Decomposition (2025 · 25 citations)
   Scientific hypothesis generation has had no dedicated benchmark for LLMs. ResearchBench addresses this by building on a decomposition from MOOSE-Chem. In that view, a hypothesis comes from combining a research background with one or more inspirations. Discovery can then be split into three sub-tasks: finding…

All works

4 works of theirs in our library, the most cited first; 2 with a page here, the others link their original until we summarise them.

1. Feature Erasing and Diffusion Network for Occluded Person Re-Identification ↗IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · no summary yetZhikang Wang et al.2022 · 207 citations
2. Online Pseudo Label Generation by Hierarchical Cluster Dynamics for Adaptive Person Re-identification ↗IEEE/CVF International Conference on Computer Vision (ICCV) · no summary yetYi Zheng et al.2021 · 113 citations
3. Many Heads Are Better Than One: Improved Scientific Idea Generation by A LLM-Based Multi-Agent SystemAnnual Meeting of the Association for Computational Linguistics (ACL)Haoyang Su, Renqi Chen, Shixiang Tang, et al.2025 · 40 citations
4. ResearchBench: Benchmarking LLMs in Scientific Discovery via Inspiration-Based Task DecompositionarXiv:2503.21248Yujie Liu, Zonglin Yang, Tong Xie, et al.2025 · 25 citations

🤝 Co-authors here

12 of 16 co-authors with a page, the most papers together first.

• Wanli Ouyang ×2
• Ben Gao
• Biqing Qi
• Bowen Zhou
• Dongzhan Zhou
• Erik Cambria
• Hui Li
• Jinjie Ni
• Jinzhe Li
• Philip H. S. Torr
• Qi Wu
• Tong Xie

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

• Annual Meeting of the Association for Computational Linguistics (ACL): ACL 2025
• IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings
• IEEE/CVF International Conference on Computer Vision (ICCV)

🔗 Cited by

Papers of our library cite their works 51 times (of the 149,439 papers whose bibliographies we read); 20 citing papers with a page here →.

Updated 3 November 2026

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/authors/shixiang-tang/