Home

Learning for Embodied Minds and Agents at the National University of Singapore.

Lab

Learning for Embodied Minds and Agents

Welcome to the LEMA Lab in the Department of Computer Science at the National University of Singapore. We are a collaborative group of researchers working to build intelligent agents that can perceive, reason, and act in the physical world. Our research bridges machine learning, computer vision, and robotics, with a focus on learning agents that can generalize from limited data, adapt during deployment, remember what matters, and improve through experience.

Open Positions

We welcome PhD students, postdoctoral researchers, and visiting students interested in building the next generation of embodied agents, with research directions including robot learning, world models, multimodal agents, data curation, and recursive self-improvement.

Join Us Hiring from 2027 All levels welcome

Research Directions

We study how embodied agents can learn from interaction, reason about the physical world, and improve autonomously.

Adaptive Robot Policies

How can robot policies adapt dynamically at inference time to handle new tasks, environments, and embodiment variations?

Predictive World Models

How can structured world models help agents anticipate spatial, temporal, and physical interactions before acting?

Proactive Self-Improvement

How can agents verify and learn from their own predictions to improve continually with minimal human supervision?

Selected Publications

Below is a representative selection of recent publications. For the full publication list, please refer to the Google Scholar profile.

World Action Verifier: Self-Improving World Models via Forward-Inverse Asymmetry

Yuejiang Liu, Fan Feng, Lingjing Kong, Weifeng Lu, Jinzhou Tang, Kun Zhang, Kevin Murphy, Chelsea Finn, Yilun Du

Preprint, 2026. Best Paper, ICLR World Model Workshop

Paper Project Code

RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies

Yinpei Dai, Hongze Fu, Jayjun Lee, Yuejiang Liu, Haoran Zhang, Jianing Yang, Chelsea Finn, Nima Fazeli, Joyce Chai

International Conference on Machine Learning (ICML), 2026. Oral (0.7%)

Paper Project Code

Learning Long-Context Diffusion Policies via Past-Token Prediction

Marcel Torne, Andy Tang, Yuejiang Liu, Chelsea Finn

Conference on Robot Learning (CoRL), 2025. Best Paper, RSS Robot Representation Workshop

Paper Project Code

Demo-SCORE: Curating Demonstrations using Online Experience

Annie Chen, Alec Lessing, Yuejiang Liu, Chelsea Finn

Robotics: Science and Systems (RSS), 2025

Paper Project Code

Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling

Yuejiang Liu, Jubayer Ibn Hamid, Annie Xie, Yoonho Lee, Max Du, Chelsea Finn

International Conference on Learning Representations (ICLR), 2025

Paper Project Code

Team Members

Principal Investigator

Yuejiang Liu

Assistant Professor, NUS

Presidential Young Professorship

Department of Computer Science

Website Scholar Twitter

Current

Recruiting from 2027, NUS

Jacky Kwok, PhD, Stanford

Xilun Zhang, PhD, Stanford

Yinpei Dai, PhD, U Michigan

Alec Lessing, MSc, Stanford

Yi Du, MSc, Stanford

Eric Liang, MSc, Stanford

Alumni

Xilun Zhang, MSc, CMU → PhD, Stanford

Jubayer Hamid, BSc, Stanford → PhD, Stanford

Frano Rajič, MSc, EPFL → PhD, ETH

Qi Yan, MSc, EPFL → PhD, UBC

Riccardo Cadei, MSc, EPFL → PhD, ISTA

Sherwin Bahmani, MSc, TUD → PhD, U Toronto

Rhea Malhotra, BSc, Stanford (Best Thesis)

Hao Zhao, MSc, EPFL → PhD, EPFL

Danya Li, MSc, EPFL → PhD, DTU