PPO-Clip gradient equals a per-sample KL surrogate with closed-form coefficient on importance ratio and advantage, yielding identical curves on five MuJoCo tasks.
Rethinking kl regularization in rlhf: From value estimation to gradient optimization
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
method 1
citation-polarity summary
verdicts
UNVERDICTED 5representative citing papers
Introduces the first benchmark for fine-grained failures in reinforcement fine-tuning of LLMs and an automatic management framework that detects, diagnoses, and remediates them.
Skill-SD turns an agent's completed trajectories into dynamic natural-language skills that condition only the teacher in self-distillation, yielding 14-42% gains over RL and OPSD baselines on multi-turn agent benchmarks.
OpenRLHF is a new open-source RLHF framework reporting 1.22x to 1.68x speedups and fewer lines of code than prior systems.
GenEvolve introduces a self-evolving agent framework for image generation using tool-orchestrated trajectories and Visual Experience Distillation to achieve claimed SOTA results on benchmarks.
citing papers explorer
-
KLip-PPO: A per-sample KL perspective on PPO-Clip
PPO-Clip gradient equals a per-sample KL surrogate with closed-form coefficient on importance ratio and advantage, yielding identical curves on five MuJoCo tasks.
-
Skill-SD: Skill-Conditioned Self-Distillation for Multi-turn LLM Agents
Skill-SD turns an agent's completed trajectories into dynamic natural-language skills that condition only the teacher in self-distillation, yielding 14-42% gains over RL and OPSD baselines on multi-turn agent benchmarks.