English(EN) Learning to Compute Is Not Learning to Retain or Reuse in Looped Transformers

循环Transformer在超出训练深度后难以保留答案

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-15 04:00

一篇新的研究论文探讨了循环Transformer的行为，这种模型通过重复应用单个块来处理信息。研究发现，尽管这些模型在超出其训练深度的扩展循环次数下可以准确计算，但它们保留正确答案的能力却显著下降。具体来说，在循环次数32下训练的模型在循环次数64下仅保留了22%的答案。该研究还区分了计算、保留和复用，证明即使是具有高保留率的模型在状态复用方面也存在困难，这表明学习计算并不等同于有效地学习保留或复用信息。 AI

影响突出了某些Transformer架构在长上下文推理能力方面存在的潜在局限性。

排序理由关于Transformer模型行为的学术论文。[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 →

本文如何被排名

Signal score: 16 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 关于Transformer模型行为的学术论文。[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]

arXiv cs.LG TIER_1 English(EN) · Peilong Zhou, Hongtao Zhang, Yulong Huang, Cangyuan Li, Mingyu Chen, Ying Wang · 2026-10-15 04:00

学习计算并非在循环Transformer中学习保留或复用

arXiv:2610.19768v1 Announce Type: new Abstract: Looped transformers compute by applying one block repeatedly, so they can run for more loops than they were trained with. Recurrent reasoners are known to overthink: iterating past the trained depth can turn a correct answer into a …

本文如何被排名

报道来源 [1]

学习计算并非在循环Transformer中学习保留或复用

相关实体

相关话题