CoRL spotlight: test-time gradient guidance tackles RL scaling woes for flow policies
chris_j_paxton · x · 2026-11-12
Chris Paxton highlighted Zhiyuan Zhou's CoRL spotlight on test-time gradient guidance for flow policies in RL. The slides argue RL has a scaling problem — actor and critic objectives are interdependent, causing instability and hyperparameter sensitivity, and flow/diffusion policies suffer most, needing specialized training objectives or backprop through time.
More from Embodied
- Smartwatch PPG + accelerometer signals reframed to detect hypertension signs — CurieuxExplorer · 2026-11-12
- Ex-Tesla Autopilot simulation leads exit stealth at $1.1B valuation with $160M raised — heyshrutimishra · 2026-11-12
- MIMIC and BMW train video-action model hitting up to 100% success on industrial tasks — pess_r · 2026-11-12
- Simular launches box, a dedicated computer built for running AI agents — _akhaliq · 2026-11-12
- Dex hand hits 99% on rubber-band task — hardware, not AI, is the bottleneck — chris_j_paxton · 2026-11-12
- Dyna co-founder Jason Ma keynotes on scaling real-world robot foundation models — chris_j_paxton · 2026-11-12