English(EN) Where Should a Document Live: Context, Representations, or Parameters?

新研究比较LLM知识注入方法

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-16 04:00

一篇新研究论文探讨了大语言模型（LLM）在预训练数据之外访问新信息的方法。该研究在五个知识密集型基准测试中比较了基于表示（KV-cache）和基于参数（微调）的适应技术。结果表明，基于KV-cache的方法，特别是“Cartridges”，在各种存储预算下提供了更高的准确性，在Oracle设置下比参数方法高出10个百分点，并在多文档检索场景中显著优于参数方法。然而，Cartridges与完全微调一样，可能导致在控制基准测试中出现灾难性遗忘。 AI

影响这项研究为优化LLM知识集成提供了见解，有望提高知识密集型任务的效率和准确性。

排序理由一篇发布在arXiv上的研究论文，详细介绍了LLM知识注入的方法。[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: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 一篇发布在arXiv上的研究论文，详细介绍了LLM知识注入的方法。[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: 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.AI TIER_1 English(EN) · Nathana\"el Carraz Rakotonirina, Momchil Hardalov, Gonzalo Iglesias, Adri\`a de Gispert · 2026-09-16 04:00

文档应存放在何处：上下文、表示还是参数？

arXiv:2609.17346v1 Announce Type: cross Abstract: To answer questions outside of their pre-training data, large language models (LLMs) need access to new information, which can be presented in the context window as documents, encoded into the model's parameters, or injected as la…

本文如何被排名

报道来源 [1]

文档应存放在何处：上下文、表示还是参数？

相关实体

相关话题