Research focus:研究方向： Multimodal Large Language Models多模态大语言模型Agentic Text-to-Image Model智能体文生图模型
I am an Algorithm Researcher at the dots team of Xiaohongshu (RED), specializing in Multimodal Large Language Models (MLLMs) and Agentic Text-to-Image Models. I received my Ph.D. from the Department of Artificial Intelligence, School of Informatics, Xiamen University, advised by Prof. Rongrong Ji and Prof. Xiaoshuai Sun. I have published 27 papers in top-tier CCF-A/B conferences and journals (17 as first/co-first author, 3 Orals), with 1.6k+ Google Scholar citations. I serve as an Area Chair for ACM MM 2025 Workshop and as a reviewer for CVPR, NeurIPS, ICLR, ICCV, IEEE TPAMI and other venues. Feel free to reach out to me via email. I am also the core developer of the open-source External-Attention-pytorch project (12.2k+ GitHub stars). In 2026, I received offers for the flagship top-talent programs of several leading tech companies:
我是小红书（RED）dots 团队的算法研究员，研究方向为多模态大语言模型（MLLMs）与智能体文生图模型。我于厦门大学信息学院人工智能系获得博士学位，师从纪荣嵘教授与孙晓帅教授。已在 CCF-A/B 类顶级会议与期刊发表论文 27 篇（其中一作/共一 17 篇，3 篇 Oral），谷歌学术引用 1.6k+ 次。担任 ACM MM 2025 Workshop 领域主席，以及 CVPR、NeurIPS、ICLR、ICCV、IEEE TPAMI 等多个会议与期刊的审稿人。欢迎通过邮箱与我联系。同时是开源项目 External-Attention-pytorch 的核心开发者（GitHub 12.2k+ stars）。于 2026 年获得多家头部科技公司顶尖人才计划录用 offer：
小红书Xiaohongshu Red Star
腾讯Tencent 青云Qingyun
阿里Alibaba 通义阿里星Tongyi Star
字节跳动ByteDance 筋斗云Jindouyun
蚂蚁Ant 蚂蚁星Star
华为Huawei 天才少年Genius Youth
美团Meituan 北斗Beidou
小米Xiaomi 顶尖人才Top Talent
JD京东 TGT
 |
dots3-note-prev (Xiaohongshu · dots)
Preview multimodal MoE model from the dots3 family, released on Hugging Face. [Hugging Face] [GitHub] [Project Page] |
 |
dots.vlm1.inst (Xiaohongshu · dots)
Instruction-tuned multimodal large language model from the dots series, released on Hugging Face. [Hugging Face] [GitHub] |
 |
dots.mocr (Xiaohongshu · dots)
Multilingual document layout parsing and OCR model from the dots series, released on Hugging Face. [Hugging Face] [GitHub] [arXiv] |
 |
Yiwei Ma, Ke Ye, Weihuang Lin, Jiayi Ji, Xiaoshuai Sun✉, Tat-Seng Chua, Rongrong Ji
An Extensive Benchmark for Single-Round and Multi-Round Instruction-Based Image Editing International Journal of Computer Vision (IJCV), 2026 [PDF] [Code] |
 |
Yiwei Ma, Jiayi Ji, Zhipeng Qian, Xiaoshuai Sun✉, Rongrong Ji
CoP: Chain of Perception for Referring 3D Instance Segmentation International Journal of Computer Vision (IJCV), 2026 [PDF] [Code] |
 |
Yiwei Ma, Weihuang Lin, Zhibin Wang, Jiayi Ji, Xiaoshuai Sun✉, Chia-Wen Lin, Rongrong Ji
Boosting Multi-Modal Large Language Model with Enhanced Visual Features IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2026 [PDF] [Code] |
 |
Yiwei Ma, Jiayi Ji, Xiaoshuai Sun✉, Yiyi Zhou, Xiaopeng Hong, Yongjian Wu, Rongrong Ji
Image Captioning via Dynamic Path Customization IEEE Transactions on Neural Networks and Learning System (TNNLS), 2024 [PDF] [ArXiv] [Code] |
 |
Yiwei Ma, Yijun Fan, Jiayi Ji, Haowei Wang, Xiaoshuai Sun✉, Guannan Jiang, Annan Shu, Rongrong Ji
X-Dreamer: Creating High-quality 3D Content by Bridging the Domain Gap Between Text-to-2D and Text-to-3D Generation ACM Transactions on Multimedia Computing, Communications, and Applications (ToMM), 2024 [PDF] [arXiv] [Code] [Project Page] |
 |
Yiwei Ma, Jiayi Ji, Xiaoshuai Sun✉, Yiyi Zhou, Rongrong Ji
Towards Local Visual Modeling for Image Captioning Pattern Recognition (PR), 2023 (ESI Highly Cited Paper)（ESI 高被引论文） [PDF] [ArXiv] [Code] |
 |
Yiwei Ma, Jiayi Ji, Xiaoshuai Sun✉, Yiyi Zhou, Yongjian Wu, Feiyue Huang, Rongrong Ji
Knowing what it is: Semantic-enhanced Dual Attention Transformer IEEE Transactions on Multimedia (TMM), 2022 [PDF] [Code] |
 |
Jiayi Ji, Yiwei Ma (co-frist author), Xiaoshuai Sun✉, Yiyi Zhou, Yongjian Wu, Rongrong Ji
Knowing What to Learn: A Metric-oriented Focal Mechanism for Image Captioning IEEE Transactions on Image Processing (TIP), 2022 [PDF] [Code] |
 |
Yiwei Ma, Jiayi Ji, Ke Ye, Weihuang Lin, Zhibin Wang, Yonghan Zheng, Qiang Zhou, Xiaoshuai Sun✉, Rongrong Ji
I2EBench: A Comprehensive Benchmark for Instruction-based Image Editing Conference on Neural Information Processing Systems (NeurIPS), 2024 [arXiv] [Code] |
 |
Yiwei Ma, Zhekai Lin, Jiayi Ji, Yijun Fan, Xiaoshuai Sun✉, Rongrong Ji
X-Oscar: A Progressive Framework for High-quality Text-guided 3D Animatable Avatar Generation International Conference on Machine Learning (ICML), 2024 [arXiv] [Code] [Project Page] |
 |
Yiwei Ma, Xiaoqing Zhang, Xiaoshuai Sun✉, Jiayi Ji, Haowei Wang, Guannan Jiang, Weilin Zhuang, Rongrong Ji
X-Mesh:Towards Fast and Accurate Text-driven 3D Stylization via Dynamic Textual Guidance IEEE International Conference on Computer Vision (ICCV), 2023 [PDF] [arXiv] [Code] [Project Page] |
 |
Yiwei Ma, Guohai Xu, Xiaoshuai Sun✉, Ming Yan, Ji Zhang, Rongrong Ji
X-CLIP: End-to-End Multi-grained Contrastive Learning for Video-Text Retrieval ACM International Conference on Multimedia (ACM MM), 2022 （Cite: 500+） [PDF] [arXiv] [Code] [Project Page] |
 |
Yiwei Ma, Xiaoshuai Sun✉, Jiayi Ji, Guannan Jiang, Weilin Zhuang, Rongrong Ji
Beat: Bi-directional One-to-Many Embedding Alignment for Text-based Person Retrieval ACM International Conference on Multimedia (ACM MM), 2023 [PDF] [Code] [Project Page] |
 |
Weihuang Lin, Yiwei Ma (co-frist author), Xiaoshuai Sun✉, Shuting He, Jiayi Ji, Liujuan Cao, Rongrong Ji
HRSeg: High-Resolution Visual Perception and Enhancement for Reasoning Segmentation ACM International Conference on Multimedia (ACM MM), 2025 [arXiv] [Code] |
 |
Zhipeng Qian, Yiwei Ma (co-frist author), Zhekai Lin, Jiayi Ji, Xiawu Zheng, Xiaoshuai Sun✉, Rongrong Ji
Multi-branch Collaborative Learning Network for 3D Visual Grounding European Conference on Computer Vision (ECCV), 2024 [arXiv] [Code] |
 |
Sihan Liu, Yiwei Ma (co-frist author), Xiaoqing Zhang, Haowei Wang, Jiayi Ji✉, Xiaoshuai Sun, Rongrong Ji
Rotated Multi-Scale Interaction Network for Referring Remote Sensing Image Segmentation Computer Vision and Pattern Recognition Conference (CVPR), 2024 [arXiv] [Code] |
 |
Changli Wu, Yiwei Ma (co-frist author), Qi Chen, Haowei Wang, Gen Luo, Jiayi Ji✉, Xiaoshuai Sun
3D-STMN: Dependency-Driven Superpoint-Text Matching Network for End-to-End 3D Referring Expression Segmentation AAAI Conference on Artificial Intelligence (AAAI), 2024 [arXiv] [Code] [PDF] |
 |
Zhipeng Qian, Yiwei Ma (co-frist author), Jiayi Ji, Xiaoshuai Sun ✉
X-RefSeg3D: Enhancing Referring 3D Instance Segmentation via Structured Cross-Modal Graph Neural Networks AAAI Conference on Artificial Intelligence (AAAI), 2024 [Code] [PDF] |
 |
Yiwei Ma, Guohai Xu, Xiaoshuai Sun✉, Jiayi Ji, Jie Lou, Debing Zhang, Rongrong Ji
MLLM-Selector: Necessity and Diversity-driven High-value Data Selection for Enhanced Visual Instruction Tuning arXiv preprint arXiv:2503.20502, 2025 [arXiv] |
 |
Yiwei Ma, Zhibin Wang, Xiaoshuai Sun✉, Weihuang Lin, Qiang Zhou, Jiayi Ji, Rongrong Ji
INF-LLaVA: Dual-perspective Perception for High-Resolution Multimodal Large Language Model arXiv preprint arXiv:2407.16198, 2024 [arXiv] [Code] |
 |
Weihuang Lin, Yiwei Ma (co-frist author), Jiayi Ji, Xiaoshuai Sun✉, Rongrong Ji
CIR-CoT: Towards Interpretable Composed Image Retrieval via End-to-End Chain-of-Thought Reasoning arXiv preprint arXiv:2510.08003, 2025 [arXiv] |
 |
Yijun Fan, Yiwei Ma (co-frist author), Jiayi Ji, Xiaoshuai Sun✉, Rongrong Ji
Any-to-3D Generation via Hybrid Diffusion Supervision arXiv preprint arXiv:2411.14715, 2024 [arXiv] |
 |
External-Attention-pytorch Pytorch implementation of various Attention Mechanisms, MLP, Re-parameter, Convolution, which is helpful to further understand papers. [github ] (12,000+ stars) |
| 已授权 一种基于文本信息的指向性3D实例分割方法 — 公开号: CN117634486A [详情] |
| 已授权 一种基于链式感知的指向性3D实例分割方法 — 公开号: CN117593527A [详情] |
| 已授权 利用文本到图像扩散模型实现短语级定位的方法 — 公开号: CN118247799A [详情] |
| 已授权 一种面向指向性目标分割的半监督学习方法 — 公开号: CN117975241A [详情] |
| 已授权 面向局部视觉建模的图像描述生成方法 — 公开号: CN115964530A [详情] |
| 已授权 基于文本的人物检索的双向一对多嵌入对齐方法 — 公开号: CN116304145A [详情] |
| 已授权 基于空间感知网络的三维指向性目标分割方法 — 公开号: CN118365659A [详情] |
| 已授权 通过混合扩散监督生成多模态到三维物体的方法 — 授权公告号: CN119625216B [详情] |
| 已授权 基于多模态大语言模型的对话生成方法及装置 — 授权公告号: CN119938874B [详情] |