Xiaofeng Zhang

Ph.D. Student, Shanghai Jiao Tong University

Key Laboratory of System Control and Information Processing
Shanghai Jiao Tong University

📧 framebreak@sjtu.edu.cn | 💬 WeChat: SemiZxf

钱塘江上潮信来，今日方知我是我

[Google Scholar] [GitHub]

📰 News

2/2026: We have one paper (SOPE, 3D-LLM-ROPE) accepted by CVPR 2026, congratulations to Guanting Ye!
2/2026: LVLMs-Saliency has been chosen as ICLR 2026 Oral!
1/2026: Three papers accepted by ICLR 2026, congratulations!
11/2025: Two papers accepted by AAAI 2026, congratulations to Shuochen Chang and Peng Gao.
9/2025: One paper accepted by IPM (中科院一区 Top).
9/2025: One paper (Spatial-R1) accepted by NeurIPS 2025, congratulations to Yifan Shen and Yuanzhe Liu.
8/2025: One paper (EAH) accepted by EMNLP 2025 Oral 🏆, see you in Suzhou.
7/2025: One paper (MCA-LLaVA) accepted by ACM MM 2025, congratulations to Qiyan Zhao.
6/2025: One paper (pCR Prediction in Breast Cancer) accepted by MICCAI 2025, congratulations to Dingrui Ma.
6/2025: One paper (AdaToken-3D, VLM token pruning) accepted by IROS 2025, congratulations to Kai Zhang.
6/2025: One paper (Mural inpainting) accepted by ACM TOMM, congratulations to Zishan Xu.
5/2025: One paper (Image segmentation) accepted by ICML 2025, congratulations to Jiawei Cao.
4/2025: One paper (Image restoration) accepted by IJCAI 2025, congratulations to Jiesong Bai.
1/2025: One paper (LLaVA-CAM) accepted by NAACL 2025 Oral 🏆.
1/2025: One paper accepted by ICLR 2025, congratulations to Sinan Fan.
1/2025: One paper (Wakeup-Darkness) accepted by ACM TOMM.
12/2024: One paper (SimIgnore) accepted by Neural Networks.
12/2024: One paper accepted by IEEE TIM, congratulations to Jietao Yang.
12/2024: One paper accepted by AAAI 2025.
10/2024: One paper accepted by WACV 2025, congratulations to Yingtie Lei.
9/2024: One paper accepted by NeurIPS 2024, congratulations to Xiaosong Yuan.
8/2024: One paper (DOPRA) accepted by ACM MM 2024.

👤 About Me

I am currently a third-year Ph.D. student at Shanghai Jiao Tong University. Prior to that, I received my Master's degree from Nanjing University of Posts and Telecommunications. My current research focuses on large vision-language models, including hallucination mitigation, position encoding optimization, and efficient multimodal reasoning.

📚 Biography

2022.09 – Present: Ph.D. student at Shanghai Jiao Tong University, supervised by Prof. Chaochen Gu and Prof. Hao Tang.
2024.01 – 2025.07: Research Intern at Alibaba Cloud (飞天实验室), supervised by Dr. Chen Shen.
2021.09 – 2022.09: Product Manager at China Mobile Communications Group Jiangsu Co., Ltd. (Wuxi Branch).
2014.09 – 2021.06: B.Eng. & M.Eng. at Nanjing University of Posts and Telecommunications.

📄 Publications (First Author)

	Hallucination Begins Where Saliency Drops Xiaofeng Zhang, Yuanchao Zhu, Chaochen Gu, Xiaosong Yuan, Qiyan Zhao, Jiawei Cao, Feilong Tang, Sinan Fan, Yaomin Shen, Chen Shen, Hao Tang ICLR 2026 Oral 🏆 [OpenReview] [Code]

	Context Tokens are Anchors: Understanding the Repetition Curse in Diffusion MLLMs from an Information Flow Perspective Qiyan Zhao, Xiaofeng Zhang✉*, Shuochen Chang, Qianyu Chen, Xiaosong Yuan, Xuhang Chen, Luoqi Liu, Jiajun Zhang, Xu-Yao Zhang, Da-Han Wang ICLR 2026, corresponding author* [OpenReview]

	What Drives Attention Sinks? A Study of Massive Activations and Rotational Positional Encoding in Large Vision-Language Models Xiaofeng Zhang, Yuanchao Zhu, Chaochen Gu, Jiawei Cao, Hao Cheng, Kaijie Wu Information Processing & Management (中科院一区), CCF B

	Seeing Clearly by Layer Two: Enhancing Attention Heads to Alleviate Hallucination in LVLMs Xiaofeng Zhang, Yihao Quan, Chaochen Gu, Chen Shen, Xiaosong Yuan, Shaotian Yan, Jieping Ye EMNLP 2025 Oral 🏆 [Arxiv] [Code]

	From Redundancy to Relevance: Information Flow in LVLMs Across Reasoning Tasks Xiaofeng Zhang, Yihao Quan, Chen Shen, Xiaosong Yuan, Shaotian Yan, Chaochen Gu, Hao Tang, Jieping Ye NAACL 2025 Oral 🏆 [arXiv] [Code]

	Enhancing Multimodal Large Language Models Complex Reasoning via Similarity Computation Xiaofeng Zhang✉, Fanshuo Zeng, Yihao Quan, Zheng Hui, Jiawei Yao AAAI 2025 [Code]

	SimIgnore: Enhancing Multimodal Large Language Models Complex Reasoning via Similarity Computation Xiaofeng Zhang, Fanshuo Zeng, Chaochen Gu Neural Networks (中科院一区) [Code]

	Wakeup-Darkness: When Multimodal Meets Unsupervised Low-light Image Enhancement Xiaofeng Zhang, Zishan Xu, Hao Tang, Chaochen Gu, Wei Chen ACM Transactions on Multimedia Computing, Communications, 2025 [IEEE Xplore] [Code]

	Memory Augment is All You Need for Image Restoration Xiaofeng Zhang, Chaochen Gu, Shanying Zhu IEEE Transactions on Consumer Electronics (中科院二区), 2026 [arXiv] [Code]

📄 Publications (Corresponding Author / Project Lead)

	Remember Me: Bridging the Long‑Range Gap in LVLMs with Three-Step Inference-Only Decay Resilience Strategies Peng Gao, Yujian Lee, *Xiaofeng Zhang✉*, Zailong Chen, Hui Zhang AAAI 2026, corresponding author* [Arxiv]

	D3ToM: Decider-Guided Dynamic Token Merging for Accelerating Diffusion MLLMs Shuochen Chang, Xiaofeng Zhang, Qingyang Liu, Li Niu AAAI 2026, project leader [Arxiv] [Code]

	MCA-LLaVA: Manhattan Causal Attention for Reducing Hallucination in Large Vision-Language Models Qiyan Zhao, Xiaofeng Zhang✉*, Yiheng Li, Yun Xing, Xiaosong Yuan, Feilong Tang, Sinan Fan, Xuhang Chen ACM MM 2025 (corresponding author / project leader)* [Arxiv] [Code]

	DOPRA: Decoding Over-accumulation Penalization and Re-allocation in Specific Weighting Layer Jingfeng Wei, Xiaofeng Zhang✉** ACM MM 2024 🏆 (corresponding author / project leader) [arXiv]

🏆 Awards

2021 National Scholarship

🔍 Services

Invited Reviewer for:

Journals: TPAMI, TIP, TETCI, IPM, Neural Networks, IEEE TCE
Conferences: ICML, NeurIPS, CVPR, ICLR, ICCV, AAAI, IJCAI, ACM MM, ACL, EMNLP, NAACL

	Remember Me: Bridging the Long‑Range Gap in LVLMs with Three-Step Inference-Only Decay Resilience Strategies Peng Gao, Yujian Lee, *Xiaofeng Zhang✉*, Zailong Chen, Hui Zhang AAAI 2026, corresponding author* [Arxiv]

	D3ToM: Decider-Guided Dynamic Token Merging for Accelerating Diffusion MLLMs Shuochen Chang, Xiaofeng Zhang, Qingyang Liu, Li Niu AAAI 2026, project leader [Arxiv] [Code]

	MCA-LLaVA: Manhattan Causal Attention for Reducing Hallucination in Large Vision-Language Models Qiyan Zhao, Xiaofeng Zhang✉*, Yiheng Li, Yun Xing, Xiaosong Yuan, Feilong Tang, Sinan Fan, Xuhang Chen ACM MM 2025 (corresponding author / project leader)* [Arxiv] [Code]

	DOPRA: Decoding Over-accumulation Penalization and Re-allocation in Specific Weighting Layer Jingfeng Wei, Xiaofeng Zhang✉** ACM MM 2024 🏆 (corresponding author / project leader) [arXiv]