specific method · filed under post-training
Binary terminal-verifier reward
Uses a binary verifier at the end of a trajectory and propagates its reward through the trajectory for credit assignment.
Also called binary verifier.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- core
How sources treat it
One count per evidence span, weakest treatment to strongest.
core 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
long-horizon credit assignment is carried entirely by the binary verifier at the end of the trajectory, propagated to every token of the trajectory through the advantage
coretraining objectivein LAGUNA M.1/XS.2Poolside
Filed alongside
Other methods under post-training :: reward modelling.
Generative Reward ModelGroupwise Agentic GradingGroupwise Reward SynthesisAdversarial screeningLength-adjusted RL rewardVerifier cross-checkingAbstention-aware reward for factual QAAgentic Generative Reward ModelBehavior rubricsBinary task verifierCollaboration bonusDeterministic chain of checkersDirect RL optimization of a generative reward modelFive-dimension comparative grading of passing patchesHack-agent screeningHybrid reward systemLanguage consistency rewardMonitoring-only penalty strategyMulti-level reward formulationMultiplicative reward synthesisNegative checks for unintended side effectsOutcome Reward ModelPer-token tool-error reward shapingPrinciple-conditioned Generative Reward Model