Jitendra Malik pushes back on Isola: LLMs can't handle high-frequency robot control
sainingxie · x · 2026-09-08
Jitendra Malik disagrees with Phillip Isola on LLMs for robot control: locomotion needs high-frequency policies dealing with forces and contacts. He challenges Isola to implement quadruped locomotion policies (RSS 2021/CoRL 2022) in an LLM, arguing quasi-static manipulation ≠ dexterity; LLMs can do high-level planning but that's not the hard part. The linked CoRL 2022 Best Systems Paper is an end-to-end egocentric-vision quadruped locomotion policy (single front depth camera, sim-trained, zero fine-tuning transfer) crossing stairs, gaps and stepping stones.
Related event: Robotics Legend Malik Pushes Back on LLM-Driven Robot Control(2 posts)→
More from Embodied
- Dev Builds a TikTok-Style Mobile UI for 15+ Years of Replication Archives — Josikinz · 2026-09-08
- USC lab's SIMPLE lands at CoRL 2026: 60-task humanoid sim benchmark for VLA models — zhengyiluo · 2026-09-08
- Swarm Robotics Paper Tests When Sociality Pays Off: Scarce Environments Erase the Advantage — abenitezburraco · 2026-09-08
- GPT-6 Astra builds 3D site exploding a humanoid robot into 1,168 CAD parts — freelerobot · 2026-09-08
- MotionVLA accepted at CoRL 2026: remember motion, not frames — yshan2u · 2026-09-08
- Robotics researcher: harness + tool calls beat raw VLA on control benchmarks, at 2.3x lower cost — rbhar90 · 2026-09-08