Multi-turn calibration reframes LLM confidence as dynamic across conversation turns, where user feedback degrades it, and new methods MTCal and ConfChat restore calibration while improving factuality.
InFindings of the Association for Computational Linguistics ACL 2024, pages 8702–8718
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Confidence Should Be Calibrated More Than One Turn Deep
Multi-turn calibration reframes LLM confidence as dynamic across conversation turns, where user feedback degrades it, and new methods MTCal and ConfChat restore calibration while improving factuality.