Solve-Detect-Verify: Inference-Time Scaling with Flexible Generative Verifier

ACL ARR 2026 January Submission9348 Authors

Keywords: Large Language Models, LLM Reasoning, Inference Time Scaling, Test Time Scaling, Token Efficiency, Reinforcement Finetuning, Verification, Thinking Fast and Slow

Abstract: Complex reasoning with Large Language Models (LLMs) demands a careful balance between accuracy and computational cost. Verification is crucial for reliability but faces trade-off: robust process-based verifiers are computationally prohibitive, while fast verifiers lack precision. We introduce flexive, a unified generative verifier designed to navigate this trade-off by dynamically allocating compute between rapid fast thinking and deliberative slow thinking. A key innovation is our training strategy: we use Group Relative Policy Optimization (GRPO) to specifically enhance the reliability of the fast mode. This targeted training generalizes effectively, elevating the slow mode to state-of-the-art open-source performance. To deploy flexive, we propose the solve-detect-verify (SDV) pipeline. Moving beyond static Best-of-N ranking, SDV employs an iterative refinement process that utilizes likelihood-based probing to detect solution completion, curtailing overthinking, and leverages flexive's feedback for targeted correction. Solve-detect-verify establishes a new open-source state-of-the-art on ProcessBench, outperforming GenPRM-32B while requiring ~2.3x fewer TFLOPS and 15x less training data. On AIME 2024, the full SDV pipeline achieves 83.3% accuracy, surpassing strong baselines while using significantly fewer tokens.

Paper Type: Long

Research Area: Sentiment Analysis, Stylistic Analysis, and Argument Mining

Research Area Keywords: Mathematical reasoning, LLM Efficiency, Chain-of-thought, Reinforcement learning, Math QA, Inference methods, Mathematical NLP

Contribution Types: NLP engineering experiment, Approaches low compute settings-efficiency

Languages Studied: English

Submission Number: 9348