{
  "id": 5240719,
  "title": "Language Models Can Control Their Own Attention",
  "url": "https://urgent.news/2026/09/02/language-models-can-control-their-own-attention",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-02T15:43:38.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2609.02737v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "Language models spend most of their attention on a small fraction of context, yet they read the entire KV cache to find the few tokens that matter. If the user asks about a previous detail in a 1M-token conversation, global attention layers must scan the full context to generate each token of the reply. A prominent approach mitigates this cost by pre-selecting relevant tokens via lightweight…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}