Select and Improve: Understanding the Mechanics of Post-Training for Reasoning

• URL: http://arxiv.org/abs/2606.13125v1
• Date: Thu, 11 Jun 2026 09:51:35 GMT
• Title: Select and Improve: Understanding the Mechanics of Post-Training for Reasoning
• Abstract summary: Reinforcement learning has rapidly emerged as a key component in the training of reasoning and coding models.<n>We study how and through what underlying processes capabilities are acquired or enhanced via reinforcement learning post-training.
• Score: 33.83246106196113
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reinforcement learning has rapidly emerged as a key component in the training of reasoning and coding models, yet it remains poorly understood from a mechanistic perspective. We study how and through what underlying processes capabilities are acquired or enhanced via reinforcement learning post-training. Our analysis, based on controlled math reasoning experiments with Qwen-2.5-1.5B, reveals two core mechanisms: strategy selection and strategy improvement. Our results highlight the role of SFT data and reinforcement learning data in activating these mechanisms, in particular showing how supervising the model on diverse reasoning strategies can enable strategy selection and how increasing difficulty in reinforcement learning data can enable strategy improvement. Taken together, our results provide mechanistic insight into RL training and suggest practical interventions to continue scaling reasoning capabilities.

Related papers

• What is Missing from AI Post-Training AI: An Empirical Analysis [74.56398070411755]
  We argue that this picture conflates two distinct capabilities: execution-level capability and strategy-level capability.<n>We find that across different tasks, the agent's training strategy is locked in at the very beginning.<n>We then examine three natural explanations--missing experience, missing guidance, and insufficient reasoning--with escalating interventions.
  arXiv  Detail & Related papers  (2026-08-19T16:17:39Z)
• Improved Training Mechanism for Reinforcement Learning via Online Model Selection [25.377745631929557]
  We study the problem of online model selection in reinforcement learning, where the selector has access to a class of reinforcement learning agents and learns to adaptively select the agent with the right configuration.<n>Our goal is to establish the improved efficiency and performance gains achieved by integrating online model selection methods into reinforcement learning training procedures.
  arXiv  Detail & Related papers  (2025-12-01T21:25:46Z)
• Rethinking the Role of Dynamic Sparse Training for Scalable Deep Reinforcement Learning [58.533203990515034]
  Scaling neural networks has driven breakthrough advances in machine learning, yet this paradigm fails in deep reinforcement learning (DRL)<n>We show that dynamic sparse training strategies provide module-specific benefits that complement the primary scalability foundation established by architectural improvements.<n>We finally distill these insights into Module-Specific Training (MST), a practical framework that exploits the benefits of architectural improvements and demonstrates substantial scalability gains across diverse RL algorithms without algorithmic modifications.
  arXiv  Detail & Related papers  (2025-10-14T03:03:08Z)
• Emergent Hierarchical Reasoning in LLMs through Reinforcement Learning [56.496001894673235]
  Reinforcement Learning (RL) has proven highly effective at enhancing the complex reasoning abilities of Large Language Models (LLMs)<n>Our analysis reveals that puzzling phenomena like aha moments", length-scaling'' and entropy dynamics are not disparate occurrences but hallmarks of an emergent reasoning hierarchy.
  arXiv  Detail & Related papers  (2025-09-03T18:52:49Z)
• CHBench: A Cognitive Hierarchy Benchmark for Evaluating Strategic Reasoning Capability of LLMs [10.29314561183905]
  Game-playing ability serves as an indicator for evaluating the strategic reasoning capability of large language models.<n>We propose CHBench, a novel evaluation framework inspired by the cognitive hierarchy models from behavioral economics.
  arXiv  Detail & Related papers  (2025-08-16T07:10:26Z)
• Scaling DRL for Decision Making: A Survey on Data, Network, and Training Budget Strategies [66.83950068218033]
  Scaling Laws demonstrate that scaling model parameters and training data enhances learning performance.<n>Despite its potential to improve performance, the integration of scaling laws into deep reinforcement learning has not been fully realized.<n>This review addresses this gap by systematically analyzing scaling strategies in three dimensions: data, network, and training budget.
  arXiv  Detail & Related papers  (2025-08-05T08:03:12Z)
• Beyond Accuracy: Dissecting Mathematical Reasoning for LLMs Under Reinforcement Learning [93.00629872970364]
  Reinforcement learning (RL) has become the dominant paradigm for improving the performance of language models on complex reasoning tasks.<n>We introduce SPARKLE, a fine-grained analytic framework to dissect the effects of RL across three key dimensions.<n>We study whether difficult problems -- those yielding no RL signals and mixed-quality reasoning traces -- can still be effectively used for training.
  arXiv  Detail & Related papers  (2025-06-05T07:53:59Z)
• DYSTIL: Dynamic Strategy Induction with Large Language Models for Reinforcement Learning [27.336254612018404]
  Reinforcement learning from expert demonstrations has long remained a challenging research problem.<n>Existing state-of-the-art methods using behavioral cloning plus further RL training often suffer from poor generalization, low sample efficiency, and poor model interpretability.<n>We propose a novel strategy-based reinforcement learning framework integrated with large language models (LLMs) to overcome these limitations.
  arXiv  Detail & Related papers  (2025-05-06T05:53:09Z)
• How Difficulty-Aware Staged Reinforcement Learning Enhances LLMs' Reasoning Capabilities: A Preliminary Experimental Study [16.441081996257576]
  This paper presents a rigorous experimental investigation into how difficulty-aware staged reinforcement learning strategies can substantially improve reasoning performance.<n>We show that strategically selecting training data according to well-defined difficulty levels markedly enhances RL optimization.<n>We will open-source our datasets on GitHub and Hugging Face.
  arXiv  Detail & Related papers  (2025-04-01T14:18:38Z)
• On the Diminishing Returns of Complex Robust RAG Training in the Era of Powerful LLMs [85.688901949146]
  We investigate the question: does the benefit of complex robust training methods diminish as language models become more powerful?<n>Our analysis reveals a consistent trend: emphthe marginal robustness benefit of sophisticated training strategies decreases substantially as model capacity increases.<n>Further investigation demonstrates that stronger models naturally exhibit better confidence calibration, cross-dataset generalization capability, and more effective attention patterns, even under simple training regimes.
  arXiv  Detail & Related papers  (2025-02-17T03:34:31Z)
• Variance-Covariance Regularization Improves Representation Learning [28.341622247252705]
  We adapt a self-supervised learning regularization technique to supervised learning contexts, introducing Variance-Covariance Regularization (VCReg) We demonstrate that VCReg significantly enhances transfer learning for images and videos, achieving state-of-the-art performance across numerous tasks and datasets. In summary, VCReg offers a universally applicable regularization framework that significantly advances transfer learning and highlights the connection between gradient starvation, neural collapse, and feature transferability.
  arXiv  Detail & Related papers  (2023-06-23T05:01:02Z)
• Rethinking Learning Dynamics in RL using Adversarial Networks [79.56118674435844]
  We present a learning mechanism for reinforcement learning of closely related skills parameterized via a skill embedding space. The main contribution of our work is to formulate an adversarial training regime for reinforcement learning with the help of entropy-regularized policy gradient formulation.
  arXiv  Detail & Related papers  (2022-01-27T19:51:09Z)
• Transfer Learning in Deep Reinforcement Learning: A Survey [64.36174156782333]
  Reinforcement learning is a learning paradigm for solving sequential decision-making problems. Recent years have witnessed remarkable progress in reinforcement learning upon the fast development of deep neural networks. transfer learning has arisen to tackle various challenges faced by reinforcement learning.
  arXiv  Detail & Related papers  (2020-09-16T18:38:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.