arXiv AI By Fernando Diaz

Offline Preference-Based Trajectory Evaluation

Read the original on arXiv AI →

arXiv:2606. 17541v1 Announce Type: cross Abstract: Offline evaluation of agentic systems often collapses trajectories to terminal success, discarding information about partial progress and inducing widespread ties, creating substantial statistical inefficiency by reducing effective sample size and weakening the ability to distinguish systems.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

Hugging Face Trending Papers
Jun 22

Self-Evolution for Multi-Turn Tool-Calling Agents via Divergence-Point Preference Learning

Multi-turn tool-using agents must coordinate long-horizon tool sequences while tracking dialogue state and policy constraints. Existing approaches often separate inference-time orchestration from parameter-level learning, leaving tool selection weakly structured and preference updates vulnerable to train--deployment prompt mismatch.