Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models
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_ | 1866912923811577856 |
|---|---|
| author | Venkatraman, Siddarth Jain, Vineet Mittal, Sarthak Shah, Vedant Obando-Ceron, Johan Bengio, Yoshua Bartoldson, Brian R. Kailkhura, Bhavya Lajoie, Guillaume Berseth, Glen Malkin, Nikolay Jain, Moksh |
| author_facet | Venkatraman, Siddarth Jain, Vineet Mittal, Sarthak Shah, Vedant Obando-Ceron, Johan Bengio, Yoshua Bartoldson, Brian R. Kailkhura, Bhavya Lajoie, Guillaume Berseth, Glen Malkin, Nikolay Jain, Moksh |
| contents | Test-time scaling methods improve the capabilities of large language models (LLMs) by increasing the amount of compute used during inference to make a prediction. Inference-time compute can be scaled in parallel by choosing among multiple independent solutions or sequentially through self-refinement. We propose Recursive Self-Aggregation (RSA), a test-time scaling method inspired by evolutionary methods that combines the benefits of both parallel and sequential scaling. Each step of RSA refines a population of candidate reasoning chains through aggregation of subsets to yield a population of improved solutions, which are then used as the candidate pool for the next iteration. Empirically, RSA delivers substantial performance gains with increasing compute budgets across diverse tasks, model families and sizes. Notably, RSA with Gemini 3 Flash attains performance near the top of the ARC-AGI-2 public leaderboard. RSA also enables Qwen3-4B-Instruct-2507 to achieve competitive performance with larger reasoning models, including DeepSeek-R1 and o3-mini (high), outperforming purely parallel and sequential scaling strategies across AIME-25, HMMT-25, Reasoning Gym, LiveCodeBench-v6, and SuperGPQA. We further propose a novel aggregation-aware reinforcement learning approach that yields significant performance gains by training the model to combine solutions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_26626 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models Venkatraman, Siddarth Jain, Vineet Mittal, Sarthak Shah, Vedant Obando-Ceron, Johan Bengio, Yoshua Bartoldson, Brian R. Kailkhura, Bhavya Lajoie, Guillaume Berseth, Glen Malkin, Nikolay Jain, Moksh Machine Learning Test-time scaling methods improve the capabilities of large language models (LLMs) by increasing the amount of compute used during inference to make a prediction. Inference-time compute can be scaled in parallel by choosing among multiple independent solutions or sequentially through self-refinement. We propose Recursive Self-Aggregation (RSA), a test-time scaling method inspired by evolutionary methods that combines the benefits of both parallel and sequential scaling. Each step of RSA refines a population of candidate reasoning chains through aggregation of subsets to yield a population of improved solutions, which are then used as the candidate pool for the next iteration. Empirically, RSA delivers substantial performance gains with increasing compute budgets across diverse tasks, model families and sizes. Notably, RSA with Gemini 3 Flash attains performance near the top of the ARC-AGI-2 public leaderboard. RSA also enables Qwen3-4B-Instruct-2507 to achieve competitive performance with larger reasoning models, including DeepSeek-R1 and o3-mini (high), outperforming purely parallel and sequential scaling strategies across AIME-25, HMMT-25, Reasoning Gym, LiveCodeBench-v6, and SuperGPQA. We further propose a novel aggregation-aware reinforcement learning approach that yields significant performance gains by training the model to combine solutions. |
| title | Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2509.26626 |