Multi-modaL Learning Lab

MLL Lab
Machine Learning and Language

We develop intelligent language and multimodal models that reason, plan, and interact with the physical world. By integrating vision, robotics, and other modalities, we build trustworthy foundation models with spatial intelligence and compositional reasoning.

Join Us Announcing RAGEN: Training RL Agents - Github ⭐️ 1.4k

📰 Latest News

  • March 2025 — Will organize the CVPR 2025 Workshop on Foundation Models for Embodied Agents, with Ruohan Zhang, Yunzhu Li, Jiayuan Mao and Wenlong Huang, Qineng Wang, Yonatan Bisk, Shenlong Wang, Fei-Fei Li and Jiajun Wu. Calling for papers!

  • December 2024 — Will organize the tutorial on Foundation Models for Embodied Agents at AAAI 2025 and NAACL 2025, with Yunzhu Li, Jiayuan Mao and Wenlong Huang.

  • December 2024 — Prof. Manling Li is selected as AAAI 2025 New Faculty Highlights.

  • November 2024 — Our Embodied Agent Interface has been selected as Best Paper Award (top 0.4%) at SoCal NLP 2024!

  • October 2024 — Our Embodied Agent Interface has been selected as Oral Presentation (top 0.6% over D&B track, top 0.4% over all tracks) at NeurIPS 2024!