Yiran Qin

MA-VLA achieves robust multi-arm compositional generalization, outperforming existing models by effectively enabling role-agnostic behavior in unseen collaboration scenarios.

MemLearner achieves unprecedented scene consistency in video generation by learning to query context memory, outperforming traditional methods in dynamic and occluded environments.

Single-view RGB input can revolutionize how robots perceive and manipulate transparent objects, achieving reliable grasping without complex depth sensing.

Generating robot training data that bridges the sim2real gap doesn't require painstakingly detailed simulation environments; instead, a neural simulator can transform classical simulations into realistic representations using only a small amount of real-world data.

Stop hard-coding reasoning strategies for your LLM agent: a learned router that dynamically picks the best paradigm for each task boosts performance by up to 5.5%, beating even the best fixed strategy.

Robots can now adapt to unforeseen errors and dynamically replan trajectories in real-time by simply incorporating sparse, human-provided or planner-provided "referring points" into their visuomotor policies.

Coordinating embodied multi-agent systems doesn't require end-to-end training; instead, offload planning to a VLM in simulation and transfer back to the real world for execution.