English(EN) CRoS: Causality Reward on Scrambled CoTs Improves Accuracy and Reasoning Faithfulness

新的CRoS指标通过打乱思维链步骤来增强LLM推理

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-02 05:00

研究人员引入了CRoS，一种用于评估语言模型思维链（CoT）推理因果关系的新指标。与之前扰动推理步骤的方法不同，CRoS在不改变内容的情况下打乱了这些步骤的顺序，旨在更好地分离真正的因果关系。这种方法旨在防止结果奖励可能带来的信息量减少或自我纠正等问题。在使用Gemma 3-4B模型在Reasoning Gym上进行的评估中，CRoS在准确性、CoT质量和忠实度方面均优于现有的因果关系和结果奖励。 AI

影响改进了对LLM推理的评估，可能带来更可靠和忠实的AI系统。

排序理由学术论文，介绍了一种评估LLM推理的新颖方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

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: Tool; 学术论文，介绍了一种评估LLM推理的新颖方法。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; 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: 1 days old; Coverage has settled into its steady-state source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CL TIER_1 English(EN) · Xiao Yu, Justin Chih-Yao Chen, Zaid Khan, Hyunji Lee, Mohit Bansal · 2026-11-02 05:00

CRoS: Causality Reward on Scrambled CoTs Improves Accuracy and Reasoning Faithfulness

arXiv:2610.42054v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has become the standard method for training language models to reason. However, RLVR typically relies solely on outcome correctness, which does not guarantee that the CoT causall…

本文如何被排名

报道来源 [1]

CRoS: Causality Reward on Scrambled CoTs Improves Accuracy and Reasoning Faithfulness

相关实体

相关话题