ACM MM 2026
Han Fang (方瀚)
I am currently a Research Scientist at the Xingchen AGI Lab, China Telecom Artificial Intelligence Technology (Beijing) Co., Ltd., where I focus on developing generative AI and advancing its applications in downstream domains. As a core contributor, I led the research and development of the 'Xinghe' Vision Model series, the text-to-image model TeleImage-1.0, and the multimodal large model TeleMM-2.0.
My research aims to bridge the gap between vision and language by enabling models to perform vision-centric reasoning, with a focus on long video understanding, free-form multimodal grounding, and vision-language pretraining.
I am open to academic collaborations and welcome motivated research interns. Please feel free to reach out at fanghan1996@outlook.com.
方瀚
我目前在中电信人工智能科技（北京）有限公司星辰 AGI 实验室工作，主要关注生成式人工智能及其在实际场景中的落地应用。作为核心研发成员，我参与并主导了 “星河”视觉模型系列、文生图模型 TeleImage-1.0 和多模态大模型 TeleMM-2.0 的研发。
我的研究围绕视觉与语言的深度融合展开，重点关注模型在视觉信息上的理解、定位与推理能力。近期方向包括长视频理解、开放式多模态定位和视觉-语言预训练。
欢迎对相关方向感兴趣的同学和研究者交流合作，也欢迎对科研感兴趣的同学联系我：fanghan1996@outlook.com。
News
- 2026.10One paper is accepted by NeurIPS 2026.
- 2026.07Two papers are accepted by ACM MM 2026.
- 2026.01TeleMM-2.0-Thinking ranked 2nd on the OpenCompass Multi-modal Academic Leaderboard.
- 2025.11One paper is accepted by AAAI 2026.
- 2025.10Ranked 3rd in the ICCV 2025 Multimodal Large Language Model Visual Reasoning Localization Challenge MARS2: VG-RS.
- 2025.04One paper is accepted by TAFFC 2025.
- 2024.12One paper is accepted by AAAI 2025.
- 2024.07One paper is accepted by ACM MM 2024.
- 2024.03Two papers are accepted by ICME 2024 (oral presentation).
- 2023.11Served as a key Contributor to the release of TeleImage 1.0, part of the 'Xingchen' Model Family, at Tianyi Expo 2023.
- 2023.07Two papers are accepted by ACM MM 2023.
- 2023.04Served as a key Contributor to the release of the 'Xinghe' Vision Model 2.0 at the China Telecom Cloud Ecosystem Conference 2023.
- 2022.12One paper is accepted by TMM 2022.
- 2022.07Joined China Telecom Artificial Intelligence Technology (Beijing) Co., Ltd. as a research scientist, focusing on text-to-image models, video-language models and multimodal large language models.
- 2021.03–2021.09Research internship at PCG, Tencent, Beijing.
- 2020.12–2021.02Research internship at MIG, SenseTime, Beijing.
最新动态
- 2026.10一篇论文被 NeurIPS 2026 接收。
- 2026.07两篇论文被 ACM MM 2026 接收。
- 2026.01TeleMM-2.0-Thinking 在 OpenCompass 多模态学术榜单中排名第二。
- 2025.11一篇论文被 AAAI 2026 接收。
- 2025.10在 ICCV 2025 多模态大语言模型视觉推理定位挑战赛 MARS2: VG-RS 中获得第三名。
- 2025.04一篇论文被 TAFFC 2025 接收。
- 2024.12一篇论文被 AAAI 2025 接收。
- 2024.07一篇论文被 ACM MM 2024 接收。
- 2024.03两篇论文被 ICME 2024 接收，并作口头报告。
- 2023.11作为核心贡献者参与发布“星辰”模型家族中的 TeleImage 1.0，2023 天翼数字科技生态大会。
- 2023.07两篇论文被 ACM MM 2023 接收。
- 2023.04作为核心贡献者参与发布“星河”视觉模型 2.0，2023 中国电信云生态大会。
- 2022.12一篇论文被 TMM 2022 接收。
- 2022.07加入中电信人工智能科技（北京）有限公司，研究方向包括文生图模型、视频-语言模型和多模态大语言模型。
- 2021.03-2021.09腾讯 PCG 北京实习。
- 2020.12-2021.02商汤科技 MIG 北京实习。
Publications
ACM MM 2026
AAAI 2026
AAAI 2025
TMM 2022
arXiv preprint 2026MeSD: Multi-Evidence Self-Distillation for VideoLLM, Weijie Zhu, Han Fang, Hanyu Fu, Yuzhe Zhang, Xin Wei, Zhaoyan Pan, Feiran Liu, Xunjie Jin, Hongbo Sun, Zhiyu Lin, Tianyi Gao, Tianyi Ding, Ye Yuan, Zhongjiang He, Hao Sun*, Zhiheng Wu*.
NeurIPS 2026GEAR-Align: Grounding-Evidence-Aware Gradient Routing for Multimodal Alignment, Yongkang Yu, Haobo Wang, Meng Chen, Han Fang, Xin Wei, Zhiyu Lin, Ye Yuan, Chiming Duan, Hao Sun, Haiyang Zhang*.
ACM MM 2026Mixture-of-Thought-Tokens: Unifying Perception and Reasoning for Free-form Multimodal Grounding, Tianyi Gao, Han Fang, Tianyi Ding, Hao Li, Xin Wei, Hongbo Sun, xiaodong dong, Ye Yuan, Jinglin Xu, Kongming Liang, Hao Sun* and Jingmin Xin*.ACM MM 2026GeoAnchor: Collaborative Reasoning via Latent Decomposition for 3D Spatial Understanding, Hao Li, Han Fang, Zixin Pan, Xin Wei, Hongbo Sun, Jinglin Xu, Zhiyu Lin, Ye Yuan, Zhongjiang He, Yu Yu* and Hao Sun*.arXiv preprint 2026SSVP: Synergistic Semantic-Visual Prompting for Industrial Zero-Shot Anomaly Detection, Chenhao Fu, Han Fang, Xiuzheng Zheng, Wenbo Wei, Yonghua Li* and Hao Sun*.arXiv preprint 2026Structure-Aware Prototype Guided Trusted Multi-View Classification, Haojian Huang, Jiahao Shi, Zhe Liu, Harold Haodong Chen, Han Fang, Hao Sun* and Zhongjiang He*.AAAI 2026Adaptive Evidential Learning for Temporal-Semantic Robustness in Moment Retrieval, Haojian Huang, Kaijing Ma, Jin Chen, Haodong Chen, Zhou Wu, Xianghao Zang, Han Fang, Chao Ban, Hao Sun*, Mulin Chen* and Zhongjiang He.ICDAR 2025SelectVision: Adaptive Vision Resolution Selection for Visual Document Understanding, Zhongjiang He, An Zhao, Ye Yuan, Han Fang, Hao Sun, Kongming Liang and Zhanyu Ma*.TAFFC 2025DDL: Dynamic Direction Learning for Semi-Supervised Facial Expression Recognition, Yaqi Li, Jing Jiang, Yuhang Zhang, Han Fang, Jiani Hu and Weihong Deng*.AAAI 2025Trusted Unified Feature-Neighborhood Dynamics for Multi-View Classification, Haojian Huang, Chuanyu Qin, Zhe Liu, Kaijing Ma, Jin Chen, Han Fang, Chao Ban, Hao Sun* and Zhongjiang He*.arXiv preprint 2025Bovila: Bootstrapping Video-Language Alignment via LLM-Based Self-Questioning and Answering, Jin Chen, Kaijing Ma, Haojian Huang, Jiayu Shen, Han Fang, Xianghao Zang, Chao Ban, Zhongjiang He, Hao Sun* and Yanmei Kang*.ICASSP 2025FASTER: Face Attribute Sliders with Semantic Rewards, Jingyan Chen, Lanxiang Zhou, Han Fang, Zerun Feng, Chao Ban, Yaqi Li, Hao Sun* and Jiani Hu*.ICASSP 2025ViCo: A Multitask Video-Enhanced and Cognition-Preserving Modality Alignment Training Framework, Zhenda Yu, Jin Chen, Jiayu Shen, Lanxiang Zhou, Han Fang, Xianghao Zang, Chao Ban, Jingfen Chen, Zhongjiang He, Hao Sun, Zerui Li* and Yanmei Kang*.ACM MM 2024GOAL: Grounded Text-to-Image Synthesis with Joint Layout Alignment Tuning, Yaqi Li, Han Fang, Zerun Feng, Kaijing Ma, Chao Ban, Xianghao Zang, Lanxiang Zhou, Zhongjiang He, Jingyan Chen, Jiani Hu*， Hao Sun* and Huayu Zhang.ICME 2024ProTA: Probabilistic Token Aggregation for Text-Video Retrieval (Oral), Han Fang, Xianghao Zang, Chao Ban, Zerun Feng, Lanxiang Zhou, Zhongjiang He, Yongxiang Li and Hao Sun*.ICME 2024Disentangle and Denoise: Tackling Context Misalignment for Video Moment Retrieval (Oral), Kaijing Ma, Han Fang, Xianghao Zang, Chao Ban, Lanxiang Zhou, Zhongjiang He, Yongxiang Li, Hao Sun, Zerun Feng and Xingsong Hou*.ACM MM 2023A Baseline Investigation: Transformer-Based Cross-View Baseline for Text-Based Person Search, Xianghao Zang, Wei Gao*, Ge Li, Han Fang, Chao Ban, Zhongjiang He and Hao Sun*.ACM MM 2023Mask to Reconstruct: Cooperative Semantics Completion for Video-Text Retrieval, Han Fang, Zhifei Yang, Xianghao Zang, Chao Ban， Zhongjiang He, Hao Sun* and Lanxiang Zhou.TMM 2022Transferring Image-CLIP to Video-Text Retrieval via Temporal Relations, Han Fang, Pengfei Xiong*, Luhui Xu and Wenhan Luo.TMM 2021Dynamic Training Data Dropout for Robust Deep Face Recognition, Yaoyao Zhong, Weihong Deng*, Han Fang, Jiani Hu, Dongyue Zhao, Xian Li and Dongchao Wen.ICASSP 2021FASTER: Face Attribute Sliders with Semantic Rewards, Hongyu Chen, Ruifang Liu*, Han Fang and Ximing Zhang.ECCV 2020Generate to Adapt: Resolution Adaptation Network for Surveillance Face Recognition, Han Fang, Weihong Deng*, Yaoyao Zhong and Jiani Hu.
论文发表
ACM MM 2026
ACM MM 2026
AAAI 2026
AAAI 2025
TMM 2022
arXiv preprint 2026MeSD: Multi-Evidence Self-Distillation for VideoLLM, Weijie Zhu, Han Fang, Hanyu Fu, Yuzhe Zhang, Xin Wei, Zhaoyan Pan, Feiran Liu, Xunjie Jin, Hongbo Sun, Zhiyu Lin, Tianyi Gao, Tianyi Ding, Ye Yuan, Zhongjiang He, Hao Sun*, Zhiheng Wu*.NeurIPS 2026GEAR-Align: Grounding-Evidence-Aware Gradient Routing for Multimodal Alignment, Yongkang Yu, Haobo Wang, Meng Chen, Han Fang, Xin Wei, Zhiyu Lin, Ye Yuan, Chiming Duan, Hao Sun, Haiyang Zhang*.ACM MM 2026Mixture-of-Thought-Tokens: Unifying Perception and Reasoning for Free-form Multimodal Grounding, Tianyi Gao, Han Fang, Tianyi Ding, Hao Li, Xin Wei, Hongbo Sun, xiaodong dong, Ye Yuan, Jinglin Xu, Kongming Liang, Hao Sun* and Jingmin Xin*.ACM MM 2026GeoAnchor: Collaborative Reasoning via Latent Decomposition for 3D Spatial Understanding, Hao Li, Han Fang, Zixin Pan, Xin Wei, Hongbo Sun, Jinglin Xu, Zhiyu Lin, Ye Yuan, Zhongjiang He, Yu Yu* and Hao Sun*.arXiv preprint 2026SSVP: Synergistic Semantic-Visual Prompting for Industrial Zero-Shot Anomaly Detection, Chenhao Fu, Han Fang, Xiuzheng Zheng, Wenbo Wei, Yonghua Li* and Hao Sun*.arXiv preprint 2026Structure-Aware Prototype Guided Trusted Multi-View Classification, Haojian Huang, Jiahao Shi, Zhe Liu, Harold Haodong Chen, Han Fang, Hao Sun* and Zhongjiang He*.AAAI 2026Adaptive Evidential Learning for Temporal-Semantic Robustness in Moment Retrieval, Haojian Huang, Kaijing Ma, Jin Chen, Haodong Chen, Zhou Wu, Xianghao Zang, Han Fang, Chao Ban, Hao Sun*, Mulin Chen* and Zhongjiang He.ICDAR 2025SelectVision: Adaptive Vision Resolution Selection for Visual Document Understanding, Zhongjiang He, An Zhao, Ye Yuan, Han Fang, Hao Sun, Kongming Liang and Zhanyu Ma*.TAFFC 2025DDL: Dynamic Direction Learning for Semi-Supervised Facial Expression Recognition, Yaqi Li, Jing Jiang, Yuhang Zhang, Han Fang, Jiani Hu and Weihong Deng*.AAAI 2025Trusted Unified Feature-Neighborhood Dynamics for Multi-View Classification, Haojian Huang, Chuanyu Qin, Zhe Liu, Kaijing Ma, Jin Chen, Han Fang, Chao Ban, Hao Sun* and Zhongjiang He*.arXiv preprint 2025Bovila: Bootstrapping Video-Language Alignment via LLM-Based Self-Questioning and Answering, Jin Chen, Kaijing Ma, Haojian Huang, Jiayu Shen, Han Fang, Xianghao Zang, Chao Ban, Zhongjiang He, Hao Sun* and Yanmei Kang*.ICASSP 2025FASTER: Face Attribute Sliders with Semantic Rewards, Jingyan Chen, Lanxiang Zhou, Han Fang, Zerun Feng, Chao Ban, Yaqi Li, Hao Sun* and Jiani Hu*.ICASSP 2025ViCo: A Multitask Video-Enhanced and Cognition-Preserving Modality Alignment Training Framework, Zhenda Yu, Jin Chen, Jiayu Shen, Lanxiang Zhou, Han Fang, Xianghao Zang, Chao Ban, Jingfen Chen, Zhongjiang He, Hao Sun, Zerui Li* and Yanmei Kang*.ACM MM 2024GOAL: Grounded Text-to-Image Synthesis with Joint Layout Alignment Tuning, Yaqi Li, Han Fang, Zerun Feng, Kaijing Ma, Chao Ban, Xianghao Zang, Lanxiang Zhou, Zhongjiang He, Jingyan Chen, Jiani Hu*, Hao Sun* and Huayu Zhang.ICME 2024ProTA: Probabilistic Token Aggregation for Text-Video Retrieval (Oral), Han Fang, Xianghao Zang, Chao Ban, Zerun Feng, Lanxiang Zhou, Zhongjiang He, Yongxiang Li and Hao Sun*.ICME 2024Disentangle and Denoise: Tackling Context Misalignment for Video Moment Retrieval (Oral), Kaijing Ma, Han Fang, Xianghao Zang, Chao Ban, Lanxiang Zhou, Zhongjiang He, Yongxiang Li, Hao Sun, Zerun Feng and Xingsong Hou*.ACM MM 2023A Baseline Investigation: Transformer-Based Cross-View Baseline for Text-Based Person Search, Xianghao Zang, Wei Gao*, Ge Li, Han Fang, Chao Ban, Zhongjiang He and Hao Sun*.ACM MM 2023Mask to Reconstruct: Cooperative Semantics Completion for Video-Text Retrieval, Han Fang, Zhifei Yang, Xianghao Zang, Chao Ban, Zhongjiang He, Hao Sun* and Lanxiang Zhou.TMM 2022Transferring Image-CLIP to Video-Text Retrieval via Temporal Relations, Han Fang, Pengfei Xiong*, Luhui Xu and Wenhan Luo.TMM 2021Dynamic Training Data Dropout for Robust Deep Face Recognition, Yaoyao Zhong, Weihong Deng*, Han Fang, Jiani Hu, Dongyue Zhao, Xian Li and Dongchao Wen.ICASSP 2021FASTER: Face Attribute Sliders with Semantic Rewards, Hongyu Chen, Ruifang Liu*, Han Fang and Ximing Zhang.ECCV 2020Generate to Adapt: Resolution Adaptation Network for Surveillance Face Recognition, Han Fang, Weihong Deng*, Yaoyao Zhong and Jiani Hu.
Honors and Awards
- Ranked 3rd in the ICCV 2025 Multimodal Large Language Model Visual Reasoning Localization Challenge, MARS2: VG-RS, 2025.
- Beijing Excellent Graduate Award (Top 1%), 2022.
- Beijing Excellent Graduate Award (Top 1%), 2019.
- Beijing Excellent Bachelor Dissertation Award (Top 3%), 2019.
荣誉奖励
- ICCV 2025 多模态大语言模型视觉推理定位挑战赛 MARS2: VG-RS 第三名，2025。
- 北京市优秀毕业生（前 1%），2022。
- 北京市优秀毕业生（前 1%），2019。
- 北京市优秀本科毕业论文（前 3%），2019。
Education
- 2019.09–2022.06 M.S. in Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing.
- 2015.09–2019.06 B.Eng. in Telecommunications Engineering with Management, Beijing University of Posts and Telecommunications and Queen Mary University of London, Beijing.
教育经历
- 2019.09-2022.06 北京邮电大学，信息与通信工程，硕士，北京。
- 2015.09-2019.06 北京邮电大学与伦敦玛丽女王大学，电信工程及管理，工学学士，北京。