LLM-as-a-Verifier: A General-Purpose Verification Framework
2026-07-06 • Artificial Intelligence
Artificial IntelligenceComputation and LanguageMachine LearningMultiagent SystemsRobotics
AI summaryⓘ
The authors identify a new way to improve large language models (LLMs) called verification, which means checking if a solution is correct. They introduce 'LLM-as-a-Verifier,' a method that provides detailed feedback without extra training by giving continuous scores based on the model's output probabilities. This approach improves checking accuracy by using finer scoring, repeated checks, and breaking down complex criteria. They also develop a cost-effective way to pick the best answer using these scores and show strong performance on various benchmarks. Additionally, their method helps guide reinforcement learning by giving detailed feedback to improve learning efficiency.
Large Language ModelsVerificationScoring GranularityToken LogitsAgentic TasksReinforcement LearningRanking AlgorithmsBenchmarkingProbabilistic ScoringSample Efficiency
Authors
Jacky Kwok, Shulu Li, Pranav Atreya, Yuejiang Liu, Yixing Jiang, Chelsea Finn, Marco Pavone, Ion Stoica, Azalia Mirhoseini
Abstract
Scaling pre-training, post-training, and test-time compute have become the central paradigms for improving the capabilities of LLMs. In this work, we identify verification, the ability to determine the correctness of a solution, as a new scaling axis. To unlock this and demonstrate its effectiveness, we introduce LLM-as-a-Verifier, a general-purpose verification framework that provides fine-grained feedback for agentic tasks without requiring additional training. Unlike standard LM judges that prompt LLMs to produce discrete scores for candidate solutions, LLM-as-a-Verifier computes the expectation over the distribution of scoring token logits to generate continuous scores. This probabilistic formulation enables verification to scale along multiple dimensions: (1) score granularity, (2) repeated evaluation, and (3) criteria decomposition. In particular, we show that scaling the scoring granularity leads to better separation between positive and negative solutions, resulting in more calibrated comparisons. Moreover, scaling repeated evaluation and criteria decomposition consistently lead to additional gains in verification accuracy through variance and complexity reduction. We further introduce a cost-efficient ranking algorithm for selecting the best solution among candidates using the verifier's continuous scores. LLM-as-a-Verifier achieves state-of-the-art performance on Terminal-Bench V2 (86.5%), SWE-Bench Verified (78.2%), RoboRewardBench (87.4%), and MedAgentBench (73.3%). Beyond verification, the fine-grained signals from LLM-as-a-Verifier can also serve as a proxy for estimating task progress. We build an extension for Claude Code, enabling developers to monitor and improve their own agentic systems. Finally, we show that LLM-as-a-Verifier can provide dense feedback for RL, improving the sample efficiency of SAC and GRPO on robotics and mathematical reasoning benchmarks.