Deep Self-Evolving Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Zihan, Zheng, Shun, Wen, Xumeng, Wang, Yang, Bian, Jiang, Yang, Mao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908603389050880
author Liu, Zihan
Zheng, Shun
Wen, Xumeng
Wang, Yang
Bian, Jiang
Yang, Mao
author_facet Liu, Zihan
Zheng, Shun
Wen, Xumeng
Wang, Yang
Bian, Jiang
Yang, Mao
contents Long-form chain-of-thought reasoning has become a cornerstone of advanced reasoning in large language models. While recent verification-refinement frameworks have enabled proprietary models to solve Olympiad-level problems, their effectiveness hinges on strong, reliable verification and correction capabilities, which remain fragile in open-weight, smaller-scale models. This work demonstrates that even with weak verification and refinement capabilities on hard tasks, the reasoning limits of such models can be substantially extended through a probabilistic paradigm we call Deep Self-Evolving Reasoning (DSER). We conceptualize iterative reasoning as a Markov chain, where each step represents a stochastic transition in the solution space. The key insight is that convergence to a correct solution is guaranteed as long as the probability of improvement marginally exceeds that of degradation. By running multiple long-horizon, self-evolving processes in parallel, DSER amplifies these small positive tendencies, enabling the model to asymptotically approach correct answers. Empirically, we apply DSER to the DeepSeek-R1-0528-Qwen3-8B model. On the challenging AIME 2024-2025 benchmark, DSER solves 5 out of 9 previously unsolvable problems and boosts overall performance, enabling this compact model to surpass the single-turn accuracy of its 600B-parameter teacher through majority voting. Beyond its immediate utility for test-time scaling, the DSER framework serves to diagnose the fundamental limitations of current open-weight reasoners. By clearly delineating their shortcomings in self-verification, refinement, and stability, our findings establish a clear research agenda for developing next-generation models with powerful, intrinsic self-evolving capabilities.
format Preprint
id arxiv_https___arxiv_org_abs_2510_17498
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Deep Self-Evolving Reasoning
Liu, Zihan
Zheng, Shun
Wen, Xumeng
Wang, Yang
Bian, Jiang
Yang, Mao
Computation and Language
Long-form chain-of-thought reasoning has become a cornerstone of advanced reasoning in large language models. While recent verification-refinement frameworks have enabled proprietary models to solve Olympiad-level problems, their effectiveness hinges on strong, reliable verification and correction capabilities, which remain fragile in open-weight, smaller-scale models. This work demonstrates that even with weak verification and refinement capabilities on hard tasks, the reasoning limits of such models can be substantially extended through a probabilistic paradigm we call Deep Self-Evolving Reasoning (DSER). We conceptualize iterative reasoning as a Markov chain, where each step represents a stochastic transition in the solution space. The key insight is that convergence to a correct solution is guaranteed as long as the probability of improvement marginally exceeds that of degradation. By running multiple long-horizon, self-evolving processes in parallel, DSER amplifies these small positive tendencies, enabling the model to asymptotically approach correct answers. Empirically, we apply DSER to the DeepSeek-R1-0528-Qwen3-8B model. On the challenging AIME 2024-2025 benchmark, DSER solves 5 out of 9 previously unsolvable problems and boosts overall performance, enabling this compact model to surpass the single-turn accuracy of its 600B-parameter teacher through majority voting. Beyond its immediate utility for test-time scaling, the DSER framework serves to diagnose the fundamental limitations of current open-weight reasoners. By clearly delineating their shortcomings in self-verification, refinement, and stability, our findings establish a clear research agenda for developing next-generation models with powerful, intrinsic self-evolving capabilities.
title Deep Self-Evolving Reasoning
topic Computation and Language
url https://arxiv.org/abs/2510.17498