PulseAugur
中
实时 08:07:46
English(EN) Rethinking Fixed-Horizon Latent Reasoning

新研究质疑LLM中固定视野的潜在推理

研究人员发现,将大型语言模型中的潜在思维链(CoT)推理扩展到固定视野有时会降低性能。该研究识别出潜在CoT的不同阶段,其中早期计算对于推理至关重要,而后期阶段可能会重组现有信息并削弱准确性。这表明需要重新思考固定视野的推理。研究人员提出了一种估计剩余推理进度并自适应调整推理过程中潜在视野的方法,该方法在不重新训练LLM的情况下,在算术基准测试上将推理准确性提高了多达7.2个百分点,同时将计算量减少了多达29.0%。 AI

影响 这项研究可能通过动态调整推理视野,从而实现更高效、更准确的LLM推理。

排序理由 这是一篇详细介绍LLM推理相关新发现和新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

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

新研究质疑LLM中固定视野的潜在推理

本文如何被排名

Signal score
24 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
这是一篇详细介绍LLM推理相关新发现和新方法的学术论文。[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
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.LG TIER_1 English(EN) · Tianxiang Chen, Jingyuan Zhou, Weiming Yang, Kaidi Yang ·

    重新思考固定视界潜在推理

    arXiv:2611.24001v1 Announce Type: new Abstract: Existing latent reasoners typically run for a preset latent horizon during inference. Intuitively, a longer horizon should provide more opportunity for reasoning. However, in this work, we find that extending latent Chain-of-Thought…