Step-level Verifier-guided Hybrid Test-Time Scaling for Large Language Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909777956700160 |
|---|---|
| author | Chang, Kaiyan Shi, Yonghao Wang, Chenglong Zhou, Hang Hu, Chi Liu, Xiaoqian Luo, Yingfeng Ge, Yuan Xiao, Tong Zhu, Jingbo |
| author_facet | Chang, Kaiyan Shi, Yonghao Wang, Chenglong Zhou, Hang Hu, Chi Liu, Xiaoqian Luo, Yingfeng Ge, Yuan Xiao, Tong Zhu, Jingbo |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_15512 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Step-level Verifier-guided Hybrid Test-Time Scaling for Large Language Models Chang, Kaiyan Shi, Yonghao Wang, Chenglong Zhou, Hang Hu, Chi Liu, Xiaoqian Luo, Yingfeng Ge, Yuan Xiao, Tong Zhu, Jingbo Computation and Language 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. |
| title | Step-level Verifier-guided Hybrid Test-Time Scaling for Large Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2507.15512 |