Sitong Wu | 吴思彤

PhD (3^rd Year)

Department of Computer Science and Engineering
The Chinese University of Hong Kong (CUHK)

Email: stone-wu@link.cuhk.edu.hk

Biography

I am currently a 3^rd year PhD student at Computer Science & Engineering Department, The Chinese University of Hong Kong (CUHK), under the supervision of Prof. Jiaya Jia. Before that, I got the Bachelor degree in Information Engineering at Jilin University (JLU).

My current research revolves around efficient and data-centric AI, particularly exploring their applications in current hot topics, such as large language models, multi-modality models, and generative models.

Selected Publications [Google Scholar]

*: equal contribution | Click a research direction to expand / collapse.

Expand all Collapse all

LLM & Agent10

	Unlocking Token Rewards via Training-Free Reward Attribution Sitong Wu, Haoru Tan, Bin Xia, Xichen Zhang, Jingyao Li, Shaofeng Zhang, Xiaojuan Qi, Bei Yu, Jiaya Jia IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2026 paper code
	TRAC: Teacher-Guided Token Reward with Adaptive Calibration for Robust Policy Optimization Sitong Wu, Haoru Tan, Xichen Zhang, Bin Xia, Wenhu Zhang, Xiaojuan Qi, Bei Yu, Jiaya Jia Annual Meeting of the Association for Computational Linguistics (ACL), 2026 paper code
	Beyond Majority Voting: Self-Reflective Test-Time Reinforcement Learning for LLM Reasoning Sitong Wu, Haoru Tan, Xichen Zhang, Bin Xia, Shaofeng Zhang, Xiaojuan Qi, Bei Yu, Jiaya Jia International Conference on Machine Learning (ICML), 2026 paper code
	DLoFT: Gradient-Decoupled Fine-Tuning for Generalizable Long Chain-of-Thought Reasoning Sitong Wu, Haoru Tan, Jingyao Li, Shaofeng Zhang, Xiaojuan Qi, Bei Yu, Jiaya Jia Conference on Neural Information Processing Systems (NeurIPS), 2025 paper code
	Dynamic Important Example Mining for Reinforcement Finetuning Haoru Tan, Sitong Wu^✉, Yanfeng Chen, Shizhen Zhao, Yang-Tian Sun, Tianjia Liu, Chirui Chang, Shaofeng Zhang, Samm Sun, Xiuzhe Wu, Ruobing Xie, Xiaojuan Qi IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2026 paper code
	Scaf-GRPO: Scaffolded Group Relative Policy Optimization for Enhancing LLM Reasoning Xichen Zhang, Sitong Wu, Yinghao Zhu, Haoru Tan, Shaozuo Yu, Ziyi He, Jiaya Jia International Conference on Learning Representations (ICLR), 2026 paper arXiv code
	SmartSwitch: Advancing LLM Reasoning by Overcoming Underthinking via Promoting Deeper Thought Exploration Xichen Zhang, Sitong Wu, Haoru Tan, Shaozuo Yu, Yinghao Zhu, Ziyi He, Jiaya Jia arXiv: 2510.19767 arXiv code
	SpaceSkill: Hierarchical Skill Routing for Tool-Augmented Spatial Agents Yuhuai Li, Haoru Tan, Sitong Wu, Wei Yin, Xiaojuan Qi arXiv, 2026 coming soon
	SearchGym: Bootstrapping Real-World Search Agents via Cost-Effective and High-Fidelity Environment Simulation Xichen Zhang, Ziyi He, Yinghao Zhu, Sitong Wu, Shaozuo Yu, Meng Chu, Wenhu Zhang, Haoru Tan, Jiaya Jia Annual Meeting of the Association for Computational Linguistics (ACL), 2026 paper arXiv code
	StreamArena: Toward Continuous, Interactive, and Long-Horizon Agentic Streaming Video Understanding Xichen Zhang, Guankai Li, Yinghao Zhu, Shijian Wang, Sitong Wu, Shaozuo Yu, Meng Chu, Yuan Lu, Jiaya Jia arXiv: 2608.05703 arXiv code

Multi-Modality5

	SaCo Loss: Sample-wise Affinity Consistency for Vision-Language Pre-training Sitong Wu, Haoru Tan, Zhuotao Tian, Yukang Chen, Xiaojuan Qi, Jiaya Jia IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024 paper supp code
	Mixture-of-Scores: Robust Image-Text Data Valuation via Three Lines of Code Sitong Wu, Haoru Tan, Yukang Chen, Shaofeng Zhang, Jingyao Li, Bei Yu, Xiaojuan Qi, Jiaya Jia IEEE/CVF International Conference on Computer Vision (ICCV), 2025 paper supp code
	Lyra: An Efficient and Speech-Centric Framework for Omni-Cognition Zhisheng Zhong, Chengyao Wang, Yuqi Liu, Senqiao Yang, Longxiang Tang, Yuechen Zhang, Jingyao Li, Tianyuan Qu, Yanwei Li, Yukang Chen, Shaozuo Yu, Sitong Wu, Eric Lo, Shu Liu, Jiaya Jia IEEE/CVF International Conference on Computer Vision (ICCV), 2025 paper arXiv code
	VisionLeaf: Entropy-Guided Leaf-First Reasoning for Efficient and Accurate Think-with-Image Haokun Gui, Senqiao Yang, Mingkang Zhu, Meng Chu, Sitong Wu, Changsheng Lu, Zihao Wang, Zhuotao Tian, Jiaya Jia IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2026 paper supp
	RegionBLIP: A Unified Multi-modal Pre-training Framework for Holistic and Regional Comprehension Qiang Zhou, Chaohui Yu, Shaofeng Zhang, Sitong Wu, Zhibin Wang, Fan Wang arXiv: 2308.02299 arXiv code

Data-Centric3

	Data Pruning by Information Maximization Haoru Tan, Sitong Wu^✉, Wei Huang, Shizhen Zhao, Xiaojuan Qi International Conference on Learning Representations (ICLR), 2026 paper arXiv code
	Fast Data Mixture Optimization via Gradient Descent Haoru Tan, Sitong Wu, Yanfeng Chen, Jun Xia, Ruobing Xie, Bin Xia, Xingwu Sun, Xiaojuan Qi International Conference on Learning Representations (ICLR), 2026 paper arXiv code
	Data Pruning via Moving-one-Sample-out Haoru Tan, Sitong Wu, Fei Du, Yukang Chen, Zhibin Wang, Fan Wang, Xiaojuan Qi Conference on Neural Information Processing Systems (NeurIPS), 2023 arXiv code

Computer Vision4

	Pale Transformer: A General Vision Transformer Backbone with Pale-Shaped Attention Sitong Wu, Tianyi Wu, Haoru Tan, Guodong Guo Association for the Advancement of Artificial Intelligence (AAAI), 2022, Oral paper arXiv code
	Demystify Transformers & Convolutions in Modern Image Deep Networks Xiaowei Hu, Min Shi, Weiyun Wang, Sitong Wu, Linjie Xing, Wenhai Wang, Xizhou Zhu, Lewei Lu, Jie Zhou, Xiaogang Wang, Yu Qiao, Jifeng Dai arXiv: 2211.05781 arXiv code
	Semantic Diffusion Network for Semantic Segmentation Haoru Tan, Sitong Wu, Jimin Pi Conference on Neural Information Processing Systems (NeurIPS), 2022 paper arXiv code
	UniNeXt: Exploring A Unified Architecture for Vision Recognition Fangjian Lin, Jianlong Yuan, Sitong Wu, Fan Wang, Zhibin Wang ACM Multimedia Conference (ACM MM), 2023 paper arXiv code
	CATrans: Context and Affnity Transformer for Few-Shot Segmentation Shan Zhang, Tianyi Wu, Sitong Wu, Guodong Guo International Joint Conference on Artificial Intelligence (IJCAI), 2022 paper arXiv
	StructToken: Rethinking Semantic Segmentation with Structural Prior Fangjian Lin, Zhanhao Liang, Sitong Wu, Junjun He, Kai Chen, Shengwei Tian IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2023 paper arXiv code
	PRSeg: A Lightweight Patch Rotate MLP Decoder for Semantic Segmentation Fangjian Lin, Yizhe Ma, Sitong Wu, Long Yu, Shengwei Tian IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2023 paper arXiv
	AxWin Transformer: A Context-Aware Vision Transformer Backbone with Axial Windows Fangjian Lin, Yizhe Ma, Sitong Wu, Long Yu, Shengwei Tian arXiv: 2305.01280 arXiv
	Fully Transformer Networks for Semantic Image Segmentation Sitong Wu, Tianyi Wu, Fangjian Lin, Shengwei Tian, Guodong Guo arXiv*: 2106.04108 arXiv code
	Full-scale Selective Transformer for Semantic Segmentation Fangjian Lin, Sitong Wu, Yizhe Ma, Shengwei Tian Asian Conference on Computer Vision (ACCV), 2022 paper
	Ensemble Quadratic Assignment Network for Graph Matching Haoru Tan, Chuang Wang, Sitong Wu, Xuyao Zhang, Fei Yin, Chenglin Liu International Journal on Computer Vision (IJCV), 2024 arXiv
	Proxy Graph Matching with Proximal Matching Networks Haoru Tan, Chuang Wang, Sitong Wu, Tieqiang Wang, Xuyao Zhang, Chenglin Liu Association for the Advancement of Artificial Intelligence (AAAI), 2021 paper