trends.barnas.net

model · first seen 7 Oct 2026

process reward model

Also: PRM, dense rewards, token-level reward, process reward labels

A reward model that scores intermediate reasoning steps rather than only final outcomes.

Mentions per week

0819 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

group relative policy optimizationreinforcement learning with verifiable rewardsLean theorem proverautomated proof verificationtoken-level reward attributionreinforcement learning for memory agentsBayesian reward modelingmodeling drift in physics reasoningcredit assignmentmixture of experts