🐦 Twitter Post Details

Viewing enriched Twitter post

@dair_ai

Even the best reasoning models hit an accuracy collapse beyond a certain problem complexity. Giving an LRM the exact solution algorithm doesn't fix it either. This new work, BIGMAS, improves LLM agents by taking inspiration from the human brain. BIGMAS outperforms both ReAct and Tree of Thoughts across all three tasks. It organizes specialized LLM agents as nodes in a dynamically constructed directed graph, coordinated through a centralized shared workspace inspired by global workspace theory. A GraphDesigner builds task-specific agent topologies per problem, and a global Orchestrator routes decisions using the complete shared state, eliminating the local-view bottleneck of reactive approaches. Across Game24, Six Fives, and Tower of London on six frontier LLMs, including GPT-5 and Claude 4.5, BIGMAS consistently improves accuracy. The gains are largest where models struggle most: DeepSeek-V3.2 jumps from 12% to 30% on Six Fives. Paper: https://t.co/sMqUfvHAGp Learn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c

Media 1
Media 2

📊 Media Metadata

{
  "media": [
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2033919566053826696/media_0.png",
      "filename": "media_0.png"
    },
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2033919566053826696/media_1.png",
      "filename": "media_1.png"
    }
  ],
  "processed_at": "2026-03-17T15:02:40.064918",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2033919566053826696",
  "url": "https://x.com/dair_ai/status/2033919566053826696",
  "twitterUrl": "https://twitter.com/dair_ai/status/2033919566053826696",
  "text": "Even the best reasoning models hit an accuracy collapse beyond a certain problem complexity.\n\nGiving an LRM the exact solution algorithm doesn't fix it either.\n\nThis new work, BIGMAS, improves LLM agents by taking inspiration from the human brain.\n\nBIGMAS outperforms both ReAct and Tree of Thoughts across all three tasks.\n\nIt organizes specialized LLM agents as nodes in a dynamically constructed directed graph, coordinated through a centralized shared workspace inspired by global workspace theory.\n\nA GraphDesigner builds task-specific agent topologies per problem, and a global Orchestrator routes decisions using the complete shared state, eliminating the local-view bottleneck of reactive approaches.\n\nAcross Game24, Six Fives, and Tower of London on six frontier LLMs, including GPT-5 and Claude 4.5, BIGMAS consistently improves accuracy.\n\nThe gains are largest where models struggle most: DeepSeek-V3.2 jumps from 12% to 30% on Six Fives.\n\nPaper: https://t.co/sMqUfvHAGp\n\nLearn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c",
  "source": "Twitter for iPhone",
  "retweetCount": 0,
  "replyCount": 1,
  "likeCount": 5,
  "quoteCount": 0,
  "viewCount": 272,
  "createdAt": "Tue Mar 17 14:53:07 +0000 2026",
  "lang": "en",
  "bookmarkCount": 7,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2033919566053826696",
  "displayTextRange": [
    0,
    278
  ],
  "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": "",
    "location": "",
    "followers": 92185,
    "following": 1,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sun Jul 23 09:12:45 +0000 2017",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 4270,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 178,
    "statusesCount": 3029,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2033181046104105379"
    ],
    "profile_bio": {
      "description": "Democratizing AI research, education, and technologies. New AI learning portal: https://t.co/LRnpZN7L4c",
      "entities": {
        "description": {
          "hashtags": [],
          "symbols": [],
          "urls": [
            {
              "display_url": "academy.dair.ai",
              "expanded_url": "https://academy.dair.ai/",
              "indices": [
                80,
                103
              ],
              "url": "https://t.co/LRnpZN7L4c"
            }
          ],
          "user_mentions": []
        },
        "url": {
          "urls": [
            {
              "display_url": "dair.ai",
              "expanded_url": "https://www.dair.ai/",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/lkqPZtMU5s"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/LVIXHjw5JI",
        "expanded_url": "https://twitter.com/dair_ai/status/2033919566053826696/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": [
              {
                "h": 86,
                "w": 86,
                "x": 88,
                "y": 1373
              }
            ]
          },
          "orig": {
            "faces": [
              {
                "h": 86,
                "w": 86,
                "x": 88,
                "y": 1373
              }
            ]
          }
        },
        "id_str": "2033919562543206405",
        "indices": [
          279,
          302
        ],
        "media_key": "3_2033919562543206405",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARw57xDjGxAFCgACHDnvEbRa4IgAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHDnvEOMbEAUKAAIcOe8RtFrgiAAA",
            "media_key": "3_2033919562543206405"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HDnvEOMbEAUE_Sj.png",
        "original_info": {
          "focus_rects": [
            {
              "h": 802,
              "w": 1432,
              "x": 0,
              "y": 0
            },
            {
              "h": 1432,
              "w": 1432,
              "x": 0,
              "y": 0
            },
            {
              "h": 1632,
              "w": 1432,
              "x": 0,
              "y": 0
            },
            {
              "h": 1776,
              "w": 888,
              "x": 0,
              "y": 0
            },
            {
              "h": 1776,
              "w": 1432,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1776,
          "width": 1432
        },
        "sizes": {
          "large": {
            "h": 1776,
            "w": 1432
          }
        },
        "type": "photo",
        "url": "https://t.co/LVIXHjw5JI"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2603.15371",
        "expanded_url": "https://arxiv.org/abs/2603.15371",
        "indices": [
          958,
          981
        ],
        "url": "https://t.co/sMqUfvHAGp"
      },
      {
        "display_url": "academy.dair.ai",
        "expanded_url": "https://academy.dair.ai/",
        "indices": [
          1034,
          1057
        ],
        "url": "https://t.co/LRnpZN7L4c"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "article": null
}