标题： AlgBench：大推理模型对算法的了解程度有多深？ 显示英文标题

标题： AlgBench: To What Extent Do Large Reasoning Models Understand Algorithms?

摘要： 推理能力已成为大型推理模型（LRMs）发展中的核心焦点。 尽管在MATH500和LiveCodeBench等几个推理基准上取得了显著进展，但现有的算法推理基准仍然有限，无法回答一个关键问题：LRMs是否真正掌握了算法推理？ 为了解答这个问题，我们提出了AlgBench，这是一个由专家精心挑选的基准，它在以算法为中心的范式下评估LRMs。 AlgBench包含超过3000个原始问题，涵盖了27种算法，由ACM算法专家构建，并按照一个全面的分类体系进行组织，包括欧几里得结构、非欧几里得结构、非优化、局部优化、全局优化和启发式优化类别。 对领先LRMs（例如Gemini-3-Pro、DeepSeek-v3.2-Speciale和GPT-o3）的实证评估揭示了显著的性能异质性：虽然模型在非优化任务上的表现良好（最高达92%），但在动态规划等全局优化算法上的准确率急剧下降至约49%。 进一步分析发现了\textbf{战略过度转移}，在此情况下，由于必要的低熵标记，模型过早地放弃了正确的算法设计。 这些发现暴露了以问题为中心的强化学习的根本局限性，并突显了采用以算法为中心的训练范式对于强大算法推理的必要性。 显示英文摘要

摘要： Reasoning ability has become a central focus in the advancement of Large Reasoning Models (LRMs). Although notable progress has been achieved on several reasoning benchmarks such as MATH500 and LiveCodeBench, existing benchmarks for algorithmic reasoning remain limited, failing to answer a critical question: Do LRMs truly master algorithmic reasoning? To answer this question, we propose AlgBench, an expert-curated benchmark that evaluates LRMs under an algorithm-centric paradigm. AlgBench consists of over 3,000 original problems spanning 27 algorithms, constructed by ACM algorithmic experts and organized under a comprehensive taxonomy, including Euclidean-structured, non-Euclidean-structured, non-optimized, local-optimized, global-optimized, and heuristic-optimized categories. Empirical evaluations on leading LRMs (e.g., Gemini-3-Pro, DeepSeek-v3.2-Speciale and GPT-o3) reveal substantial performance heterogeneity: while models perform well on non-optimized tasks (up to 92%), accuracy drops sharply to around 49% on globally optimized algorithms such as dynamic programming. Further analysis uncovers \textbf{strategic over-shifts}, wherein models prematurely abandon correct algorithmic designs due to necessary low-entropy tokens. These findings expose fundamental limitations of problem-centric reinforcement learning and highlight the necessity of an algorithm-centric training paradigm for robust algorithmic reasoning.