Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Retrieval-Augmented Linguistic Calibration

About

Linguistic cues such as "I believe" and "probably" offer an intuitive interface for communicating confidence, yet a generalisable, principled calibration framework for linguistic confidence expressions remains underexplored. In particular, co-occurring linguistic cues, contextual variation, and subjective audience interpretation pose unique challenges. We therefore model linguistic confidence as a distribution over plausible perceived probability values that a statement is correct, capturing interpretation variability that scalar representations discard. Within this distributional framework, we introduce faithfulness as a complementary evaluation dimension and present Faithfulness Divergence (FD), an information-theoretic metric quantifying the surprise induced in audience beliefs upon truth revelation. Building on these foundations, we present Retrieval-Augmented Linguistic Calibration (RALC), a lightweight post-hoc pipeline that propagates calibrated confidence signals back into natural language via retrieval-augmented rewriting. Across three QA benchmarks and five LLM families, RALC improves in-domain faithfulness and calibration up to 66% and 58%, respectively, outperforming black-box and grey-box calibration baselines.

Yi-Fan Yeh, Linwei Tao, Minjing Dong, Tao Huang, Jialin Yu, Philip Torr, Chang Xu• 2026

Related benchmarks

TaskDatasetResultRank
CalibrationMMLU--
58
Question AnsweringTruthfulQA
AUROC0.664
14
Linguistic-space calibrationSQuAD 2.0
Mean Faithfulness Divergence Reduction66.4
9
Linguistic-space calibrationTruthfulQA
Faithfulness Divergence Reduction70.4
9
Question AnsweringMMLU
AUROC (MMLU QA)0.636
5
Question AnsweringSQuAD 2.0
AUROC58.8
5
Showing 6 of 6 rows

Other info

Follow for update