Last updated: 14 November 2026
I am a second-year Ph.D. student in Computer Science at Dartmouth College. I am a member of the SAHAS Lab, where I am honored to be advised by Prof. Nikhil Singh and build agentic AI algorithms and systems.
Before Dartmouth, I received my M.Sc. in Data Science from The Chinese University of Hong Kong, Shenzhen, and my B.Eng. in Biomedical Engineering from Shanghai Jiao Tong University.
I worked as a research intern at Alibaba DAMO Academy, mentored by Dr. Tian Zhou. I also served as a Visiting Student with Prof. Ming-Hsuan Yang at UC Merced and Google DeepMind and Prof. Chao Ma at Shanghai Jiao Tong University. I have previously collaborated with Prof. Lu Qi and Dr. Xiangtai Li at TikTok, as well as with Prof. Junwei Liang at AI Thrust, HKUST(GZ).
My Ph.D. research builds agentic AI algorithms and systems, along three directions:
- (1) Long-Horizon Multimodal Agent Memory
- (2) Long-Horizon Agentic RL
- (3) Agentic AI System Security
I am open to research collaborations on all three directions, and I am looking for a Summer 2027 internship in the U.S. on AI agents. If either resonates with your work, feel free to email me at yujin.tang.gr@dartmouth.edu. I also run a WeChat discussion group on Agents, Multimodal Agents and Agentic RL — add me at ChrisEve1999 if you would like to join.
Educations
Internships
News
- 2026.11: 🌎🌎 We released Always-On Personal Intelligence: A Roadmap Towards Continuing, Self-Evolving Loops, a survey and roadmap on continuing, self-evolving agents. Paper Website Reading List
- 2026.09: 🎉🎉 One paper on MLLM interpretability is accepted by NeurIPS 2026. Congratulations to Chenming! Paper
- 2026.08: 🎉🎉 DMV-Bench is accepted by EMNLP 2026.
- 2026.07: 🚀🚀 DMV-Bench is online, the first interactive benchmark for multimodal-agent visual memory. Paper Code HuggingFace
- 2026.06: 🎉🎉 PDR / PredGS is accepted by ECCV 2026. Paper Code
- 2026.06: 📄📄 Our survey The Landscape of Agentic Time Series Systems: Architectures, Reliability, and Frontiers is released! See the PDF in the repository
.
- 2026.04: 📚📚 Yujin created Awesome-Agent-Memory-Papers
, a curated list of papers on agent memory research. A companion website provides organized and filterable browsing of the papers. Also synced to my blog. Hope it helps! - 2026.01: 🎉🎉 One paper is accepted by ICLR 2026.
- 2026.01: 🎉🎉 PredFormer is accepted by TMLR 2026.
- 2025.09: 🎓🎓 Yujin enrolled in Dartmouth College as a Ph.D. student in Computer Science.
- 2025.03: 🔬🔬 Yujin joined Alibaba DAMO Academy as a research intern.
- 2024.10: 🎉🎉 PredFormer is available! This is the first pure-transformer based model in spatial-temporal predictive learning, which is recurrent-free and convolution-free, outperforming previous models by large margins with superior efficiency. Paper Code
- 2024.04: 🎉🎉 VMRNN is accepted by CVPRW 2024.
- 2024.03: 🎉🎉 PostRainBench is accepted by ICLRW 2024.
- 2024.01: 🎉🎉 Yujin created Awesome-Mamba-Papers repository. This is the first Awesome Mamba repository and promotes Mamba related research. Link
Building Agentic AI Intelligence
I build agentic AI algorithms and systems.
Project: Long-Horizon Multimodal Agent Memory
More coming soon.
Project: Long-Horizon Agentic RL
More coming soon.
Project: Agentic AI System Security
More coming soon.
Survey Collaborations
Earlier Work: Predicting the Future
Before moving to agents, I worked on predicting how the world unfolds — spatiotemporal and time series forecasting.
As first author: VMRNN (CVPR Workshop 2024) paired Vision Mamba with LSTM for spatiotemporal forecasting; PredFormer (TMLR 2026) showed a pure transformer can beat recurrent and convolutional video prediction models; PredGS (ECCV 2026) learned video dynamics through differentiable rendering. I also built PostRainBench (ICLR Workshop 2024) for precipitation forecasting.
As collaborator: distribution-aware alignment for time series forecasting (ICLR 2026), PatchMixer (IJCAI Workshop 2024), and early work on proteomics-based cancer diagnosis (MICCAI 2020).
Publications
* indicates equal contribution.
Selected First-Author Publications (full list)
Selected Collaborative Publications (full list)
Community Resource
- Awesome-Agent-Memory-Papers (2026, 240+ stars) — a curated, continuously updated collection of agent memory research spanning episodic, semantic, procedural and multimodal memory, across methods, benchmarks and surveys, with a companion filterable website.
- Awesome-Mamba-Papers (2024, 1.4K stars) — the first Mamba paper collection.
- PredFormer (158 stars) and VMRNN-PyTorch (167 stars) — official implementations.
Reviewing
Agent venues: ICLR 2026 Workshop on MemAgents x3 · COLM 2026 Workshop on Agent Behavior x2
- 2026: International Journal of Computer Vision (IJCV) x2, IEEE Transactions on Multimedia (TMM) x3
- 2025: Transactions on Machine Learning Research (TMLR), International Journal of Computer Vision (IJCV)
- 2024: IEEE Transactions on Neural Networks and Learning Systems (TNNLS)
Teaching
- Spring/Summer/Fall 2026: Teaching Assistant for CS50: Software Design and Implementation, Dartmouth College
- Winter 2026: Head Teaching Assistant for Principles of Machine Learning, Dartmouth College
- Fall 2025: Teaching Assistant for Video Understanding, Dartmouth College
Miscs
- Born and raised in Dazhou, Sichuan, China. A first-generation undergraduate and Ph.D. in my family.
- Now based in Hanover, NH. Before that I lived in Dazhou, Mianyang, Shanghai, Shenzhen, Guangzhou and Hangzhou. Shenzhen is my favorite.
- I love thinking and writing. “The limits of my language mean the limits of my world.” / 语言的边界就是思想的边界。
- 100k+ followers on Chinese social media. Learning by sharing.
- Looking for startup co-founders (CTO, MTS) and a life partner. Email me if you are interested.