CROP：面向推理轨迹前缀的共形认证方法

View PDF HTML (experimental)

Abstract:Language model reasoning traces are rarely all-or-nothing; they frequently contain valid intermediate steps before a critical error occurs. Existing uncertainty quantification methods typically certify final answers or entire responses, failing to provide statistical guarantees for the proportion of a sequential trace that can be safely retained. To address this, we introduce CROP (Conformal Reasoning Output Prefixes), a verifier-agnostic calibration procedure for clean-prefix certification. Given any step-level risk proxy, CROP selects a calibrated threshold and returns the longest contiguous prefix whose step risk proxies remain below it, routing the uncertified suffix for downstream review or repair. Assuming exchangeability, CROP rigorously controls the marginal probability that the returned prefix contains an annotated error. Across six process-labeled reasoning datasets, we demonstrate that standard step-level metrics such as AUROC do not fully capture prefix utility, suggesting verifiers should instead be evaluated by certified prefix length. Furthermore, CROP balances over- and under-withholding and improves downstream repair accuracy by preserving valid intermediate reasoning while discarding misleading suffixes. Ultimately, this work positions prefix certification as a rigorous, practical bridge between process supervision, abstention, and repair. Code available at this https URL.

Comments:	Accepted at EMNLP 2026 (Oral, Main)
Subjects:	Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Machine Learning (stat.ML)
Cite as:	arXiv:2605.30085 [cs.AI]
	(or arXiv:2605.30085v2 [cs.AI] for this version)
	https://doi.org/10.48550/arXiv.2605.30085 arXiv-issued DOI via DataCite

Submission history

From: Matt Cheung [view email]
[v1] Thu, 28 May 2026 15:31:42 UTC (604 KB)
[v2] Sat, 10 Oct 2026 00:21:28 UTC (1,280 KB)