🐦 Twitter Post Details

Viewing enriched Twitter post

@dair_ai

// Language Models Need Sleep // Let your agents "sleep", folks. On a serious note, this is a fascinating paper on getting the most from long-horizon agents. Here is the problem with agents today: Attention scales badly with context length, so long-horizon agents keep paying a quadratic tax at inference time. This work proposes a sleep-like consolidation step instead. The model periodically does N offline recurrent passes over recent context, writes the result into persistent fast weights in its state-space blocks, then clears the KV cache. The effect is that extra compute moves to sleep while wake-time prediction stays low latency. On cellular automata, multi-hop graph retrieval, and a math reasoning task where a plain transformer and SSM-attention hybrids fail, longer sleep durations improve performance, with the biggest gains on examples that need deeper reasoning. Why does it matter? It points at an alternative to ever-larger KV caches for agents that run for a long time. Consolidate, then forget the raw tokens. Paper: https://t.co/FfDTbhl98M Learn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c

Media 1

📊 Media Metadata

{
  "media": [
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2059333792775745619/media_0.jpg",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2026-06-16T11:17:47.352114",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2059333792775745619",
  "url": "https://x.com/dair_ai/status/2059333792775745619",
  "twitterUrl": "https://twitter.com/dair_ai/status/2059333792775745619",
  "text": "// Language Models Need Sleep //\n\nLet your agents \"sleep\", folks.\n\nOn a serious note, this is a fascinating paper on getting the most from long-horizon agents.\n\nHere is the problem with agents today: Attention scales badly with context length, so long-horizon agents keep paying a quadratic tax at inference time.\n\nThis work proposes a sleep-like consolidation step instead. The model periodically does N offline recurrent passes over recent context, writes the result into persistent fast weights in its state-space blocks, then clears the KV cache.\n\nThe effect is that extra compute moves to sleep while wake-time prediction stays low latency. On cellular automata, multi-hop graph retrieval, and a math reasoning task where a plain transformer and SSM-attention hybrids fail, longer sleep durations improve performance, with the biggest gains on examples that need deeper reasoning.\n\nWhy does it matter?\n\nIt points at an alternative to ever-larger KV caches for agents that run for a long time. Consolidate, then forget the raw tokens.\n\nPaper: https://t.co/FfDTbhl98M\n\nLearn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c",
  "source": "Twitter for iPhone",
  "retweetCount": 45,
  "replyCount": 11,
  "likeCount": 241,
  "quoteCount": 11,
  "viewCount": 84980,
  "createdAt": "Tue May 26 18:00:11 +0000 2026",
  "lang": "en",
  "bookmarkCount": 187,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2059333792775745619",
  "displayTextRange": [
    0,
    280
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "dair_ai",
    "url": "https://x.com/dair_ai",
    "twitterUrl": "https://twitter.com/dair_ai",
    "id": "889050642903293953",
    "name": "DAIR.AI",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1643277398522187778/31dedbLo_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/889050642903293953/1773242460",
    "description": "Democratizing AI research, education, and technologies. Learn about AI Agents for FREE at https://t.co/HHXg8rryu4",
    "location": "",
    "followers": 126150,
    "following": 1,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sun Jul 23 09:12:45 +0000 2017",
    "entities": {
      "description": {
        "urls": [
          {
            "display_url": "academy.dair.ai/courses/elemen…",
            "expanded_url": "https://academy.dair.ai/courses/elements-of-ai-agents",
            "indices": [
              90,
              113
            ],
            "url": "https://t.co/HHXg8rryu4"
          }
        ]
      },
      "url": {
        "urls": [
          {
            "display_url": "dair.ai",
            "expanded_url": "https://www.dair.ai/",
            "indices": [
              0,
              23
            ],
            "url": "https://t.co/lkqPZtMU5s"
          }
        ]
      }
    },
    "fastFollowersCount": 0,
    "favouritesCount": 4593,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 245,
    "statusesCount": 3316,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2066548421670941068"
    ],
    "profile_bio": {},
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.x.com/JjVvfMfUN9",
        "expanded_url": "https://x.com/dair_ai/status/2059333792775745619/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": []
          },
          "medium": {
            "faces": []
          },
          "orig": {
            "faces": []
          },
          "small": {
            "faces": []
          }
        },
        "id_str": "2059333788489121793",
        "indices": [
          281,
          304
        ],
        "media_key": "3_2059333788489121793",
        "media_results": {
          "result": {
            "media_key": "3_2059333788489121793"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HJQ5K3jagAEe1UN.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 903,
              "w": 1612,
              "x": 0,
              "y": 0
            },
            {
              "h": 1612,
              "w": 1612,
              "x": 0,
              "y": 0
            },
            {
              "h": 1822,
              "w": 1598,
              "x": 0,
              "y": 0
            },
            {
              "h": 1822,
              "w": 911,
              "x": 319,
              "y": 0
            },
            {
              "h": 1822,
              "w": 1612,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1822,
          "width": 1612
        },
        "sizes": {
          "large": {
            "h": 1822,
            "resize": "fit",
            "w": 1612
          },
          "medium": {
            "h": 1200,
            "resize": "fit",
            "w": 1062
          },
          "small": {
            "h": 680,
            "resize": "fit",
            "w": 602
          },
          "thumb": {
            "h": 150,
            "resize": "crop",
            "w": 150
          }
        },
        "type": "photo",
        "url": "https://t.co/JjVvfMfUN9"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2605.26099",
        "expanded_url": "https://arxiv.org/abs/2605.26099",
        "indices": [
          1047,
          1070
        ],
        "url": "https://t.co/FfDTbhl98M"
      },
      {
        "display_url": "academy.dair.ai",
        "expanded_url": "https://academy.dair.ai/",
        "indices": [
          1123,
          1146
        ],
        "url": "https://t.co/LRnpZN7L4c"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "article": null
}