FAER: Auditable Utility-Aligned Trajectory Replay for Language Model Post-Training

Excerpt

arXiv:2610.00385v1 Announce Type: new Abstract: Replay selectors often rank cached trajectories by format feedback, confidence, freshness, or response length, although cache-level correctness and downstream learner utility are distinct objectives. We formalize this selection-to-learning gap and introduce FAER as an auditable full-trajectory replay framework. Its training-free fixed selector is a protocol baseline; FAER-UTILITY is the learner-aware selector fitted on disjoint calibration blocks.