🐦 Twitter Post Details

Viewing enriched Twitter post

@dair_ai

New research from IBM Research on Self-Improving Agents. Agents have "amnesia." An agent that struggles with a particular API authentication flow today will struggle with the same flow tomorrow unless manually updated. This paper introduces a framework for automatically extracting actionable learnings from agent execution trajectories and using them to improve future performance through contextual memory retrieval. The system generates three types of guidance: strategy tips from successful patterns, recovery tips from failure handling, and optimization tips from inefficient but successful executions. A Trajectory Intelligence Extractor performs semantic analysis of agent reasoning patterns while a Decision Attribution Analyzer traces backwards through reasoning steps to identify root causes. On the AppWorld benchmark, the memory-enhanced agent achieves 73.2% task goal completion compared to 69.6% baseline (+3.6 pp) and 64.3% scenario goal completion compared to 50.0% (+14.3 pp). The benefits scale with task complexity. Difficulty 3 tasks show the most dramatic improvements: +28.5 pp on scenario goals (19.1% to 47.6%), a 149% relative increase. Why it matters: Agents that learn from their own execution traces, not just from training data, can systematically improve without manual prompt engineering. The self-reinforcing cycle of better tips producing better trajectories producing better tips is a practical path toward self-improving agent systems. Paper: https://t.co/8IOIeEgFM5 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/2032459951306866714/media_0.png",
      "filename": "media_0.png"
    },
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2032459951306866714/media_1.png",
      "filename": "media_1.png"
    }
  ],
  "processed_at": "2026-03-13T14:17:12.992370",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2032459951306866714",
  "url": "https://x.com/dair_ai/status/2032459951306866714",
  "twitterUrl": "https://twitter.com/dair_ai/status/2032459951306866714",
  "text": "New research from IBM Research on Self-Improving Agents.\n\nAgents have \"amnesia.\"\n\nAn agent that struggles with a particular API authentication flow today will struggle with the same flow tomorrow unless manually updated.\n\nThis paper introduces a framework for automatically extracting actionable learnings from agent execution trajectories and using them to improve future performance through contextual memory retrieval.\n\nThe system generates three types of guidance: strategy tips from successful patterns, recovery tips from failure handling, and optimization tips from inefficient but successful executions. A Trajectory Intelligence Extractor performs semantic analysis of agent reasoning patterns while a Decision Attribution Analyzer traces backwards through reasoning steps to identify root causes.\n\nOn the AppWorld benchmark, the memory-enhanced agent achieves 73.2% task goal completion compared to 69.6% baseline (+3.6 pp) and 64.3% scenario goal completion compared to 50.0% (+14.3 pp). The benefits scale with task complexity. Difficulty 3 tasks show the most dramatic improvements: +28.5 pp on scenario goals (19.1% to 47.6%), a 149% relative increase.\n\nWhy it matters:\n\nAgents that learn from their own execution traces, not just from training data, can systematically improve without manual prompt engineering.\n\nThe self-reinforcing cycle of better tips producing better trajectories producing better tips is a practical path toward self-improving agent systems.\n\nPaper: https://t.co/8IOIeEgFM5\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": 118,
  "createdAt": "Fri Mar 13 14:13:08 +0000 2026",
  "lang": "en",
  "bookmarkCount": 2,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2032459951306866714",
  "displayTextRange": [
    0,
    273
  ],
  "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": 91676,
    "following": 1,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sun Jul 23 09:12:45 +0000 2017",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 4256,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 175,
    "statusesCount": 3014,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2032107624007876781"
    ],
    "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/bEhR6kFINP",
        "expanded_url": "https://twitter.com/dair_ai/status/2032459951306866714/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": [
              {
                "h": 81,
                "w": 81,
                "x": 817,
                "y": 1562
              }
            ]
          },
          "orig": {
            "faces": [
              {
                "h": 81,
                "w": 81,
                "x": 817,
                "y": 1562
              }
            ]
          }
        },
        "id_str": "2032459947389444096",
        "indices": [
          274,
          297
        ],
        "media_key": "3_2032459947389444096",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARw0v43AW2AACgACHDS/jqnagBoAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHDS/jcBbYAAKAAIcNL+OqdqAGgAA",
            "media_key": "3_2032459947389444096"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HDS_jcBbYAASkyE.png",
        "original_info": {
          "focus_rects": [
            {
              "h": 899,
              "w": 1606,
              "x": 0,
              "y": 0
            },
            {
              "h": 1606,
              "w": 1606,
              "x": 0,
              "y": 0
            },
            {
              "h": 1766,
              "w": 1549,
              "x": 57,
              "y": 0
            },
            {
              "h": 1766,
              "w": 883,
              "x": 397,
              "y": 0
            },
            {
              "h": 1766,
              "w": 1606,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1766,
          "width": 1606
        },
        "sizes": {
          "large": {
            "h": 1766,
            "w": 1606
          }
        },
        "type": "photo",
        "url": "https://t.co/bEhR6kFINP"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2603.10600",
        "expanded_url": "https://arxiv.org/abs/2603.10600",
        "indices": [
          1487,
          1510
        ],
        "url": "https://t.co/8IOIeEgFM5"
      },
      {
        "display_url": "academy.dair.ai",
        "expanded_url": "https://academy.dair.ai/",
        "indices": [
          1563,
          1586
        ],
        "url": "https://t.co/LRnpZN7L4c"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "article": null
}