Assistant Professor of Computer ScienceHarvard University · Kempner Institute
I study how intelligent systems can learn reusable models, then compose and reason with them at inference time.
I explore this question through four interfaces between representation and inference: search over generative solution spaces, composition of reusable models, planning with world models, and reasoning through multi-agent interaction. Across these directions, a central focus of my work is building robots that can plan, adapt, and act reliably in unfamiliar environments.
I lead the Embodied Minds Lab at Harvard and am a part-time research scientist at NVIDIA Research. I received my PhD from MIT EECS, advised by Leslie Kaelbling, Tomas Lozano-Pérez, and Joshua B. Tenenbaum. Before Harvard, I held research roles at OpenAI and Google DeepMind.
KnowledgeSolution spacesReasoningSample and refine
A generative model represents a space of possible solutions. At inference time, we sample from this space and refine candidate solutions. More inference steps enable more accurate results and the incorporation of new constraints.
Examples: deep EBMs, Reduce, Reuse, Recycle, and reasoning with sampling.
KnowledgeConcepts and skillsReasoningCombine models
Different models can represent reusable concepts, constraints, or skills. At inference time, we select and compose the models needed to solve a novel task, allowing the same learned knowledge to support many different problems.
Examples: composable diffusion models, Diffuser, and compositional energy minimization.
KnowledgeWorld dynamicsReasoningPredict and plan
A world model predicts how an environment will evolve under different actions. Planning searches these predictions for actions that achieve a goal. This enables us to use the same model across tasks and changing environments.
Examples: UniPi, Video Language Planning, and Large Video Planner.
KnowledgeDistributed across agentsReasoningDebate and verify
Language models have different knowledge and capabilities. Their interaction forms an inference procedure: agents debate, verify, and refine candidate solutions or coordinate on a shared task, enabling systematic reasoning.
Examples: multi-agent debate, multi-agent verification, and Economy of Minds.
* Equal contribution; † or + equal advising.