PulseAugur
中
实时 12:26:30
English(EN) Weaken to Strengthen: Weakening-Induced On-Policy Self-Distillation for Large Reasoning Models

新方法增强 LLM 推理和基准生成 · 跟踪 3 个来源

作者 PulseAugur 编辑部 · [3 个来源] ·

研究人员正在开发新方法来增强大型语言模型（LLM）的推理能力。一种名为 AnytimeReasoner 的方法通过适应不同的 token 预算并改进可验证的密集奖励的信用分配来优化推理效率。另一种技术，减弱诱导的策略内自蒸馏（WISD），通过创建较弱的对应模型进行对比训练来提高模型性能，而无需特权信息。此外，一个名为 DeoGen 的新基准生成器可自动创建规范推理的数据集，这对于在受监管环境中部署 LLM 至关重要。 AI

影响 LLM 推理和自动化基准测试的进步可以加速其在复杂、受监管环境中的部署。

排序理由 该集群包含三篇关于改进 LLM 推理和基准生成的不同研究论文。

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →

新方法增强 LLM 推理和基准生成 · 跟踪 3 个来源

本文如何被排名

Signal score
1 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
该集群包含三篇关于改进 LLM 推理和基准生成的不同研究论文。
Source corroboration
3 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, model release, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
3 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [3]

  1. arXiv cs.CL TIER_1 English(EN) · Penghui Qi, Zichen Liu, Tianyu Pang, Chao Du, Wee Sun Lee, Min Lin ·

    Optimizing Anytime Reasoning via Budget Relative Policy Optimization

    arXiv:2505.13438v4 Announce Type: replace-cross Abstract: Scaling test-time compute is crucial for enhancing the reasoning capabilities of large language models (LLMs). Existing approaches typically employ reinforcement learning (RL) to maximize a verifiable reward obtained at th…

  2. arXiv cs.AI TIER_1 English(EN) · Yecheng Wu, Song Han, Han Cai ·

    弱化以致增强：面向大型推理模型的弱化诱导策略内自蒸馏

    arXiv:2610.38681v1 Announce Type: new Abstract: Self-distillation offers a way to improve reasoning models when a suitable external teacher is unavailable. However, common approaches rely on privileged information to construct a stronger self-teacher, limiting their use when only…

  3. arXiv cs.AI TIER_1 English(EN) · Sam Kouteili, William Liang, Keeran Murray, Adriana Zhou, Ramesh Viswanathan, Scott Shapiro, Ruzica Piskac ·

    Deogen：策略推理的自动基准生成

    arXiv:2610.38687v1 Announce Type: new Abstract: Language models are increasingly being deployed in regulated settings, where a model must apply given rules to the facts of a situation to determine a verdict. Benchmarking this capability, known as deontic reasoning, remains a cost…

相关实体

相关话题