English(EN) Select-TTT: Query-Conditioned Selective Fast Weight Updates for In-Place Test-Time Training

新的Select-TTT方法增强了LLM长上下文问答能力

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-28 04:00

研究人员开发了Select-TTT，一种用于大型语言模型（LLM）的测试时训练（TTT）的新方法，可提高长上下文问答的性能。与之前应用统一更新的方法不同，Select-TTT根据传入文本块与用户查询的相关性来选择性地更新模型权重。这种查询条件机制使用模型自身的冻结输出头来确定相关性，无需额外的参数或监督。该方法在不同的基准测试和模型骨干上都显示出了一致的改进，表明其在涉及不断变化的上下文和多查询的代理任务中的潜力。 AI

影响增强了LLM内化和利用长上下文信息的能力，提高了复杂查询任务的性能。

排序理由研究论文，详细介绍了一种新的LLM测试时训练方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 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: 11 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 研究论文，详细介绍了一种新的LLM测试时训练方法。[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: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Yue Wu, Yang Zhang, Weinan Gan, Chongming Gao, Xinsen Guo, Xiaoyuan Li, Fuli Feng · 2026-10-28 04:00

Select-TTT：用于原地测试时训练的查询条件选择性快速权重更新

arXiv:2610.36794v1 Announce Type: cross Abstract: Test-time training (TTT) enables a deployed language model to internalize incoming context on-the-fly into a small set of fast-weight parameters, rather than merely treating it as transient context for attention. In-Place TTT rece…

本文如何被排名

报道来源 [1]

Select-TTT：用于原地测试时训练的查询条件选择性快速权重更新

相关实体

相关话题