From Apparent Search Gains to Selection Failures: Auditing Test-Time Inference in Structured Workflow Planning

arXiv:2611.00579v1 Announce Type: new Abstract: Test-time search can improve structured planning, but when methods differ in prompts, post-processing, and selection, it is unclear what an observed gain should be attributed to. We study this question on DisasterBench, where models compose typed multi

arXiv cs.CLDienstag 06:00 · 03.11.1 Min. Lesezeit