Cross-Model Disagreement as a Label-Free Correctness Signal

TL;DR

This paper introduces cross-model disagreement as a training-free, label-free method to detect when a language model is wrong, especially in confident error cases, improving deployment safety.

Contribution

The authors propose cross-model disagreement metrics, CMP and CME, that outperform traditional uncertainty measures in identifying model errors without additional training.

Findings

CMP achieves 0.75 AUROC on MMLU, outperforming baselines.

CME and CMP outperform within-model uncertainty on multiple benchmarks.

Method is applicable for deployment monitoring and model oversight.

Abstract

Detecting when a language model is wrong without ground truth labels is a fundamental challenge for safe deployment. Existing approaches rely on a model's own uncertainty -- such as token entropy or confidence scores -- but these signals fail critically on the most dangerous failure mode: confident errors, where a model is wrong but certain. In this work we introduce cross-model disagreement as a correctness indicator -- a simple, training-free signal that can be dropped into existing production systems, routing pipelines, and deployment monitoring infrastructure without modification. Given a model's generated answer, cross-model disagreement computes how surprised or uncertain a second verifier model is when reading that answer via a single forward pass. No generation from the verifying model is required, and no correctness labels are needed. We instantiate this principle as Cross-Model Perplexity (CMP), which measures the verifying model's surprise at the generating model's answer tokens, and Cross-Model Entropy (CME), which measures the verifying model's uncertainty at those positions. Both CMP and CME outperform within-model uncertainty baselines across benchmarks spanning reasoning, retrieval, and mathematical problem solving (MMLU, TriviaQA, and GSM8K). On MMLU, CMP achieves a mean AUROC of 0.75 against a within-model entropy baseline of 0.59. These results establish cross-model disagreement as a practical, training-free approach to label-free correctness estimation, with direct applications in deployment monitoring, model routing, selective prediction, data filtering, and scalable oversight of production language model systems.