Verifiable process supervision trains language models to produce accurate answers with sound, verifiable reasoning steps, outperforming accuracy-only reinforcement learning on chess by preserving accuracy while reducing reasoning errors.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Correct Answers from Sound Reasoning: Verifiable Process Supervision for Language Models
Verifiable process supervision trains language models to produce accurate answers with sound, verifiable reasoning steps, outperforming accuracy-only reinforcement learning on chess by preserving accuracy while reducing reasoning errors.