BWhat changed
Introduced response-length-adaptive advantage estimation within a value-based long-reasoning training framework, integrating value pretraining, decoupled estimators, clipping and positive-example learning.
Scientific scope
Introduced response-length-adaptive advantage estimation within a value-based long-reasoning training framework, integrating value pretraining, decoupled estimators, clipping and positive-example learning. Clip-Higher/token losses are inherited from DAPO, value pretraining/decoupled GAE from VC-PPO and sampling/self-imitation from earlier work. The specific length adaptation/integration is acquired, not invention of PPO or a universal stability theorem.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Clip-Higher/token losses are inherited from DAPO, value pretraining/decoupled GAE from VC-PPO and sampling/self-imitation from earlier work. The specific length adaptation/integration is acquired, not invention of PPO or a universal stability theorem.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction