Pith. sign in

VinePPO: Refining credit assignment in RL training of LLMs

2 Pith papers cite this work. Polarity classification is still indexing.

2 Pith papers citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.AI 1 cs.LG 1

years

2026 2

verdicts

UNVERDICTED 2

roles

background 1

polarities

background 1

representative citing papers

Verifiable Process Rewards for Agentic Reasoning

cs.AI · 2026-05-11 · unverdicted · novelty 6.0 · 2 refs

VPR converts symbolic, constraint, or posterior oracles into dense turn-level rewards for RL, improving credit assignment in agentic reasoning and transferring to general benchmarks.

citing papers explorer

Showing 2 of 2 citing papers.

  • Verifiable Process Rewards for Agentic Reasoning cs.AI · 2026-05-11 · unverdicted · none · ref 9 · 2 links

    VPR converts symbolic, constraint, or posterior oracles into dense turn-level rewards for RL, improving credit assignment in agentic reasoning and transferring to general benchmarks.

  • Beyond Uniform Credit Assignment: Selective Eligibility Traces for RLVR cs.LG · 2026-05-07 · unverdicted · none · ref 39

    S-trace adds sparse eligibility traces to RLVR that mask low-entropy tokens, outperforming GRPO by 0.49-3.16% pass@16 on Qwen3 models while improving sample and token efficiency.