English(EN) TopoCompress: Long Context Compression via Graph-Wired Semantic Trajectories

新的TopoCompress方法大幅降低LLM长上下文处理成本

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-01 04:00

研究人员开发了TopoCompress，一个旨在降低大型语言模型处理长上下文的成本和延迟的新框架。这种无需训练且与模型无关的方法通过识别和选择语义连贯的片段来实现上下文压缩。TopoCompress构建一个混合图，根据语义相似性和顺序邻近性连接这些片段，然后传播相关性分数。在五个不同的长上下文任务中，TopoCompress与现有的压缩基线相比表现更优，在显著更小的压缩预算和更快的处理时间内取得了可比的结果。 AI

影响降低处理长上下文的LLM的推理成本和延迟，可能促进此类模型的更广泛采用。

排序理由关于LLM上下文压缩新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 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: 30 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 关于LLM上下文压缩新方法的学术论文。[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, infra; 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.CL TIER_1 English(EN) · Daniel Agyei Asante, Yang Li · 2026-09-01 04:00

TopoCompress：通过图连接语义轨迹实现长上下文压缩

arXiv:2608.30811v1 Announce Type: new Abstract: Long-context compression is essential for reducing the cost and latency of large language model inference. However, existing methods can fragment important evidence, require additional training or alignment, and often depend on the …

本文如何被排名

报道来源 [1]

TopoCompress：通过图连接语义轨迹实现长上下文压缩

相关实体

相关话题