🐦 Twitter Post Details

Viewing enriched Twitter post

@dair_ai

Reasoning-Aware Retrieval for Deep Research Agents Deep research agents generate explicit reasoning before every search call. These reasoning traces encode rich signals about search intent and problem-solving context. Yet no existing retriever learns to exploit them effectively. This paper introduces AgentIR, a reasoning-aware retrieval system that jointly embeds the agent's reasoning trace alongside its query instead of just the query alone. Why does it matter? The agent's reasoning acts as a retrieval instruction, a memory of key history, and an implicit filter for outdated information. All of this context is available for free since the agent already generates it. AgentIR-4B achieves 68% accuracy on BrowseComp-Plus with the open-weight Tongyi-DeepResearch agent, compared to 52% with conventional embedding models twice its size and 37% with BM25. It also outperforms LLM-based reranking by 10% absolute, with no additional inference overhead. Paper: https://t.co/rok5nZDfYw Learn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c

View on Twitter

📊 Media Metadata

{
  "media": [
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2031726356292407366/media_0.jpg",
      "filename": "media_0.jpg"
    },
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2031726356292407366/media_1.png",
      "filename": "media_1.png"
    }
  ],
  "processed_at": "2026-03-11T13:46:50.305286",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2031726356292407366",
  "url": "https://x.com/dair_ai/status/2031726356292407366",
  "twitterUrl": "https://twitter.com/dair_ai/status/2031726356292407366",
  "text": "Reasoning-Aware Retrieval for Deep Research Agents\n\nDeep research agents generate explicit reasoning before every search call.\n\nThese reasoning traces encode rich signals about search intent and problem-solving context.\n\nYet no existing retriever learns to exploit them effectively.\n\nThis paper introduces AgentIR, a reasoning-aware retrieval system that jointly embeds the agent's reasoning trace alongside its query instead of just the query alone.\n\nWhy does it matter?\n\nThe agent's reasoning acts as a retrieval instruction, a memory of key history, and an implicit filter for outdated information. All of this context is available for free since the agent already generates it.\n\nAgentIR-4B achieves 68% accuracy on BrowseComp-Plus with the open-weight Tongyi-DeepResearch agent, compared to 52% with conventional embedding models twice its size and 37% with BM25. It also outperforms LLM-based reranking by 10% absolute, with no additional inference overhead.\n\nPaper: https://t.co/rok5nZDfYw\n\nLearn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c",
  "source": "Twitter for iPhone",
  "retweetCount": 0,
  "replyCount": 0,
  "likeCount": 2,
  "quoteCount": 0,
  "viewCount": 335,
  "createdAt": "Wed Mar 11 13:38:05 +0000 2026",
  "lang": "en",
  "bookmarkCount": 2,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2031726356292407366",
  "displayTextRange": [
    0,
    269
  ],
  "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/889050642903293953/1742055232",
    "description": "",
    "location": "",
    "followers": 91644,
    "following": 1,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sun Jul 23 09:12:45 +0000 2017",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 4242,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 171,
    "statusesCount": 3003,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2031375297506161039"
    ],
    "profile_bio": {
      "description": "Democratizing AI research, education, and technologies.",
      "entities": {
        "description": {
          "hashtags": [],
          "symbols": [],
          "urls": [],
          "user_mentions": []
        },
        "url": {
          "urls": [
            {
              "display_url": "dair.ai",
              "expanded_url": "https://www.dair.ai/",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/lkqPZtMmfU"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/WMoCTZLdGs",
        "expanded_url": "https://twitter.com/dair_ai/status/2031726356292407366/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": []
          },
          "orig": {
            "faces": []
          }
        },
        "id_str": "2031726352714604544",
        "indices": [
          270,
          293
        ],
        "media_key": "3_2031726352714604544",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARwyJFpuWqAACgACHDIkW0ObkEYAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHDIkWm5aoAAKAAIcMiRbQ5uQRgAA",
            "media_key": "3_2031726352714604544"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HDIkWm5aoAAiv5_.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 869,
              "w": 1552,
              "x": 0,
              "y": 0
            },
            {
              "h": 1552,
              "w": 1552,
              "x": 0,
              "y": 0
            },
            {
              "h": 1712,
              "w": 1502,
              "x": 0,
              "y": 0
            },
            {
              "h": 1712,
              "w": 856,
              "x": 0,
              "y": 0
            },
            {
              "h": 1712,
              "w": 1552,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1712,
          "width": 1552
        },
        "sizes": {
          "large": {
            "h": 1712,
            "w": 1552
          }
        },
        "type": "photo",
        "url": "https://t.co/WMoCTZLdGs"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2603.04384",
        "expanded_url": "https://arxiv.org/abs/2603.04384",
        "indices": [
          972,
          995
        ],
        "url": "https://t.co/rok5nZDfYw"
      },
      {
        "display_url": "academy.dair.ai",
        "expanded_url": "https://academy.dair.ai/",
        "indices": [
          1048,
          1071
        ],
        "url": "https://t.co/LRnpZN7L4c"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "article": null
}