MetaScale: Test-Time Scaling with Evolving Meta-Thoughts
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915202074673152 |
|---|---|
| author | Liu, Qin Zhou, Wenxuan Xu, Nan Huang, James Y. Wang, Fei Zhang, Sheng Poon, Hoifung Chen, Muhao |
| author_facet | Liu, Qin Zhou, Wenxuan Xu, Nan Huang, James Y. Wang, Fei Zhang, Sheng Poon, Hoifung Chen, Muhao |
| contents | One critical challenge for large language models (LLMs) for making complex reasoning is their reliance on matching reasoning patterns from training data, instead of proactively selecting the most appropriate cognitive strategy to solve a given task. Existing approaches impose fixed cognitive structures that enhance performance in specific tasks but lack adaptability across diverse scenarios. To address this limitation, we introduce METASCALE, a test-time scaling framework based on meta-thoughts -- adaptive thinking strategies tailored to each task. METASCALE initializes a pool of candidate meta-thoughts, then iteratively selects and evaluates them using a multi-armed bandit algorithm with upper confidence bound selection, guided by a reward model. To further enhance adaptability, a genetic algorithm evolves high-reward meta-thoughts, refining and extending the strategy pool over time. By dynamically proposing and optimizing meta-thoughts at inference time, METASCALE improves both accuracy and generalization across a wide range of tasks. Experimental results demonstrate that MetaScale consistently outperforms standard inference approaches, achieving an 11% performance gain in win rate on Arena-Hard for GPT-4o, surpassing o1-mini by 0.9% under style control. Notably, METASCALE scales more effectively with increasing sampling budgets and produces more structured, expert-level responses. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_13447 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MetaScale: Test-Time Scaling with Evolving Meta-Thoughts Liu, Qin Zhou, Wenxuan Xu, Nan Huang, James Y. Wang, Fei Zhang, Sheng Poon, Hoifung Chen, Muhao Computation and Language Artificial Intelligence Machine Learning One critical challenge for large language models (LLMs) for making complex reasoning is their reliance on matching reasoning patterns from training data, instead of proactively selecting the most appropriate cognitive strategy to solve a given task. Existing approaches impose fixed cognitive structures that enhance performance in specific tasks but lack adaptability across diverse scenarios. To address this limitation, we introduce METASCALE, a test-time scaling framework based on meta-thoughts -- adaptive thinking strategies tailored to each task. METASCALE initializes a pool of candidate meta-thoughts, then iteratively selects and evaluates them using a multi-armed bandit algorithm with upper confidence bound selection, guided by a reward model. To further enhance adaptability, a genetic algorithm evolves high-reward meta-thoughts, refining and extending the strategy pool over time. By dynamically proposing and optimizing meta-thoughts at inference time, METASCALE improves both accuracy and generalization across a wide range of tasks. Experimental results demonstrate that MetaScale consistently outperforms standard inference approaches, achieving an 11% performance gain in win rate on Arena-Hard for GPT-4o, surpassing o1-mini by 0.9% under style control. Notably, METASCALE scales more effectively with increasing sampling budgets and produces more structured, expert-level responses. |
| title | MetaScale: Test-Time Scaling with Evolving Meta-Thoughts |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2503.13447 |