不训练也能提升大模型推理能力，通过细粒度验证实现高效测试时扩展。
Step-level Verifier-guided Hybrid Test-Time Scaling for Large Language Models
- 用步骤级条件自修正结合过程验证，实现细粒度推理优化。
- 在3B到14B的5个模型上，推理性能显著提升。
- 适合追求高效推理且不愿训练的开发者使用。
测试时扩展（Test-Time Scaling, TTS）是一种在推理阶段逐步激发大语言模型智能的有前景方法。近期基于训练的TTS方法（如持续强化学习）愈发流行，而无训练TTS方法逐渐被忽视。然而，训练带来的额外计算开销加剧了测试时扩展的负担。本文聚焦于无训练TTS在推理任务中的应用，提出条件步骤级自修正方法，通过过程验证引导细粒度序列化扩展。在此基础上，进一步在步骤级别融合多种经典并行扩展方法，提出一种新型推理范式——混合测试时扩展。在五个不同规模（3B-14B）和架构的大语言模型上进行的广泛实验表明，以细粒度整合多种无训练TTS方法的混合策略，具有显著拓展大模型推理性能边界的能力。
原文摘要 · Abstract (English)
Test-Time Scaling (TTS) is a promising approach to progressively elicit the model's intelligence during inference. Recently, training-based TTS methods, such as continued reinforcement learning (RL), have further surged in popularity, while training-free TTS methods are gradually fading from prominence. However, the additional computation overhead of training amplifies the burden on test-time scaling. In this paper, we focus on training-free TTS methods for reasoning. We first design Conditional Step-level Self-refinement, a fine-grained sequential scaling method guided by process verification. On top of its effectiveness, we further combine it with other classical parallel scaling methods at the step level, to introduce a novel inference paradigm called Hybrid Test-Time Scaling. Extensive experiments on five instruction-tuned LLMs across different scales (3B-14B) and families demonstrate that hybrid strategy incorporating various training-free TTS methods at a fine granularity has considerable potential for expanding the reasoning performance boundaries of LLMs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。