PulseAugur
中
实时 22:46:51
English(EN) Doc-to-Atom: Learning to Compile and Compose Memory Atoms

Doc-to-Atom 框架改进 LLM 长文档推理

作者 PulseAugur 编辑部 · [2 个来源] ·

研究人员推出了一种名为 Doc-to-Atom (Doc2Atom) 的新框架，旨在改进大型语言模型处理长文档的方式。与为整个文档创建单个适配器的先前方法不同，Doc2Atom 将文档分解为“知识原子”。每个原子都被编译成一个小型、独立的适配器，可以在推理时选择性地检索和组合。这种方法旨在减少内存使用并增强长文本的推理能力，在实验中优于现有的 Doc-to-LoRA 方法。 AI

影响 提高 LLM 在处理和推理长文档方面的效率和有效性。

排序理由 这是一篇描述 LLM 新方法的论文。

在 arXiv cs.IR (Information Retrieval) 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

Doc-to-Atom 框架改进 LLM 长文档推理

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
这是一篇描述 LLM 新方法的论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
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
89 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 [2]

  1. arXiv cs.CL TIER_1 English(EN) · Xingjian Diao, Wenbo Li, Yashas Malur Saidutta, Avinash Amballa, Lazar Valkov, Srinivas Chappidi ·

    Doc-to-Atom：学习编译和组合内存原子

    arXiv:2606.12400v1 Announce Type: new Abstract: Long input sequences are central to document understanding and multi-step reasoning in Large Language Models, yet the quadratic cost of attention makes inference both memory-intensive and slow. Context distillation mitigates this by…

  2. arXiv cs.IR (Information Retrieval) TIER_1 English(EN) · Srinivas Chappidi ·

    Doc-to-Atom: 学习编译和组合内存原子

    Long input sequences are central to document understanding and multi-step reasoning in Large Language Models, yet the quadratic cost of attention makes inference both memory-intensive and slow. Context distillation mitigates this by compressing contextual information into model p…

相关实体

相关话题