RHyVE shows LLM-generated reward rankings are unreliable at low policy competence but become informative after task-dependent thresholds, with phase-aware deployment improving peak and retained performance on sparse manipulation tasks.
Table 9 summarizes the main artifact groups used to generate the paper-ready results
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
RHyVE: Competence-Aware Verification and Phase-Aware Deployment for LLM-Generated Reward Hypotheses
RHyVE shows LLM-generated reward rankings are unreliable at low policy competence but become informative after task-dependent thresholds, with phase-aware deployment improving peak and retained performance on sparse manipulation tasks.