Each Insufficient in Its Own Way: Calibrating Language Models Across Varying Evidence States
arXiv:2610.33237v1 Announce Type: new Abstract: How can a language model know whether it has enough evidence to answer? Language models tend to answer even when the evidence they have is insufficient. Deciding whether to answer or gather more evidence requires a calibrated estimate of how likely an…
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-10-27 04:00 · arXiv cs.CL
Each Insufficient in Its Own Way: Calibrating Language Models Across Varying Evidence States