Why the λ_e penalty must tangent the Pareto curve to lift the whole frontier in RL
carlesgelada · x · 2026-09-12
Follow-up in a thread on multi-effort RL reward design: λe controls the slope of iso-reward lines in the cost-performance plane, and only when those lines are tangent to the Pareto curve is improving reward guaranteed to push the entire Pareto frontier upward.
Related event: RL Reward Design: Choosing λ_e to Lift the Whole Pareto Frontier(3 posts)→
More from Research
- HABIT-Bench: Best agent memory system scores just 33.61% at inferring user habits — SJTU · 2026-10-28
- MatAnyone-E brings 7.4x faster 4K video matting via sparse memory propagation — mmlab-ntu · 2026-10-28
- CodeCap-Bench: executable code beats natural language as video captions — Jiafu Tang · 2026-10-28
- NVIDIA's ThinkWorld learns robot skills from unlabeled human videos via chain-of-dynamics reasoning — nvidia · 2026-10-28
- Omni-InteractScope benchmarks proactive interaction in continuous omnimodal streams — Qwen · 2026-10-28
- Memory Tax study: dedicated agent memory systems often lose to naive RAG — UIUC-CS · 2026-10-28