Zhenfei Yin

Achieving a 78.3% success rate in real-world mobile manipulation, this framework bridges the reality gap with zero-shot transferability to unseen tasks.

Adversarial attacks on vision-language agents reveal critical vulnerabilities, with multi-view optimization strategies proving significantly more effective than isolated approaches.

Orchestrating expert LLMs can boost scientific reasoning accuracy by over 3% while cutting API costs by more than half.

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

Current AI agents struggle to reliably rediscover scientific knowledge, with top performers averaging only 21.5 out of a possible score, revealing critical gaps in their research capabilities.

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.

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.

Imagine AI scientists that not only reason but also autonomously conduct experiments in the real world – that's the promise of Intelligent Science Laboratories.