DeepDelta Lab: Make Change (Δ) Happen!
We are DeepDelta Lab, a research group at School of Artificial Intelligence in Shanghai Jiao Tong University. The PI of DeepDelta Lab is Dr. Shaobo Cui.
We have four overarching goals:
-
Causal Intelligence: We study how machines can learn and leverage causal relationships, moving beyond correlational prediction toward intervention, counterfactual reasoning, and robust generalization. Our research includes causal representation learning, causal reasoning, causal strength modeling, and the integration of causal mechanisms with foundation models.
-
Causal World Models: We develop world models that can understand environment dynamics, predict the consequences of actions, and reason about interventions and counterfactuals. Key topics include physical and causal mechanism modeling in vision and video, long-horizon temporal reasoning, memory mechanisms, and planning and decision-making in complex environments.
-
Causality/AI for Science (Medicine & Biology): We explore foundation models and causal methods for scientific problems in medicine and biology, including multimodal biomedical data modeling, medical knowledge and literature understanding, molecular and protein modeling, uncertainty quantification, and causal reasoning and decision support for clinical applications and scientific discovery.
-
Alignment of Large Language Models: Build reliable, efficient, and steerable LLMs that can reason, plan, and collaborate. Focus particularly on the alignment stage of LLMs.
我们 (DeepDelta Lab) 是 上海交通大学人工智能学院 的一个研究组。我们的主要研究方向为：
-
因果智能（Causal Intelligence）：研究机器如何学习和利用因果关系，从相关性预测进一步走向干预、反事实推理与稳健泛化，包括因果表征学习、因果推理、因果强度建模，以及因果机制与基础模型的结合。
-
因果驱动的世界模型（Causal World Models）：研究能够理解环境动态、预测行动后果并进行干预与反事实推演的世界模型，重点关注视觉与视频中的物理和因果机制建模、长期时序推理、记忆机制，以及面向复杂环境的规划与决策。
-
Causality/AI for Science（医学与生物）：探索基础模型与因果方法在医学、生物等科学问题中的应用，包括多模态生物医学数据建模、医学知识与文献理解、分子与蛋白建模、不确定性量化，以及面向临床和科学发现的因果推理与决策支持。
-
大语言模型对齐：构建可靠、高效、可引导，且能够推理、规划与协作的 LLM，重点研究对齐阶段的关键问题。
PI 介绍 (DeepDelta Lab @ SJTU): 崔少波，上海交通大学人工智能学院长聘教轨助理教授，博士生导师，国家高层次青年人才，上海市海外高层次青年人才。博士毕业于 EPFL，获计算机与通信科学专业 博士学位，师从国际人工智能与多智能体领域知名学者 Boi Faltings 教授（AAAI Fellow、EuRAI Fellow）。 此前，他分别从北航和清华取得学士和硕士学位。崔少波博士的研究聚焦于：因果智能、因果驱动的世界模型、大语言模型、因果在医疗方向的前沿应用等。 代表性荣誉与成果包括：G-Research PhD Prize、德国 DAAD AINet Fellow (AI4Science Track)、阿里巴巴明日之星、NAACL 杰出论文奖、ACL SAC Highlight、北京市优秀毕业生、清华大学优秀学位论文等。相关技术成果已在冬奥会、阿里云等大规模场景中落地应用。作为第一开发人构建的因果工具累计下载量 超 17 万次。
Group Members
We are looking for new PhD students, Master students, Bachelor RAs to join the team (see openings) !
Faculty
Graduate Students
RAs
Follow Us
Stay connected with DeepDelta Lab for research updates, publications, opportunities, and lab news.
 Xiaohongshu |
 WeChat Official Account |
 X / Twitter |
News
- [Rolling] We are actively looking for: Postdoc, PhD student, master student, and Undergraduate intern to join our lab
- Our new work <Synthetic Persona Pretraining: Alignment from Token Zero> is online: aligning language models from very beginning!
- Our DeepDelta Lab’s paper with ETH Zurich and Google DeepMind receive attention from media. Big congratulation to Xiaochuan!
- One paper accepted at EMNLP 2026.
- One paper accepted at ICML 2026.
- One paper accepted at ACL 2026. SAC Rating: 10/10: Top 5% of accepted papers, Seminal Paper！
- One paper accepted at WWW 2026 Demo Track!
- DeepDelta Lab is officially founded!
- 2 Papers accepted at ACL 2025