🐦 Twitter Post Details

Viewing enriched Twitter post

@omarsar0

LLMs Can Learn About Themselves by Introspection This paper reports that LLMs can acquire knowledge through introspection that cannot be inferred from their training data. "Our findings challenge the view that LLMs merely imitate their training data and suggest they have privileged access to information about themselves." They also report that this introspection ability is limited and models struggle to predict their behavior on tasks requiring reasoning over long outputs. This is exciting and interesting because these introspection capabilities can lead to more interpretable and controllable LLMs.

Media 1
View on Twitter

📊 Media Metadata

{
  "data": [
    {
      "id": "",
      "type": "photo",
      "url": null,
      "media_url": "https://pbs.twimg.com/media/GaLrXtWbsAAmk4X.png",
      "media_url_https": null,
      "display_url": null,
      "expanded_url": null
    }
  ],
  "score": 1.0,
  "scored_at": "2025-08-09T13:46:07.548948",
  "import_source": "network_archive_import",
  "media": [
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1847297594525094081/media_0.png?",
      "filename": "media_0.png",
      "original_url": "https://pbs.twimg.com/media/GaLrXtWbsAAmk4X.png"
    }
  ],
  "storage_migrated": true
}

🔧 Raw API Response

{
  "user": {
    "created_at": "2015-09-04T12:59:26.000Z",
    "default_profile_image": false,
    "description": "Building with AI Agents @dair_ai • Prev: Meta AI, Elastic, Galactica LLM, PhD • I also teach how to build with LLMs, RAG & AI Agents ⬇️",
    "fast_followers_count": 0,
    "favourites_count": 27933,
    "followers_count": 216712,
    "friends_count": 532,
    "has_custom_timelines": true,
    "is_translator": false,
    "listed_count": 3690,
    "location": "",
    "media_count": 2656,
    "name": "elvis",
    "normal_followers_count": 216712,
    "possibly_sensitive": false,
    "profile_banner_url": "https://pbs.twimg.com/profile_banners/3448284313/1565974901",
    "profile_image_url_https": "https://pbs.twimg.com/profile_images/939313677647282181/vZjFWtAn_normal.jpg",
    "screen_name": "omarsar0",
    "statuses_count": 12439,
    "translator_type": "regular",
    "url": "https://t.co/JBU5beHQNs",
    "verified": true,
    "withheld_in_countries": [],
    "id_str": "3448284313"
  },
  "id": "1847297594525094081",
  "conversation_id": "1847297594525094081",
  "full_text": "LLMs Can Learn About Themselves by Introspection\n\nThis paper reports that LLMs can acquire knowledge through introspection that cannot be inferred from their training data.\n\n\"Our findings challenge the view that LLMs merely imitate their training data and suggest they have privileged access to information about themselves.\"\n\nThey also report that this introspection ability is limited and models struggle to predict their behavior on tasks requiring reasoning over long outputs.\n\nThis is exciting and interesting because these introspection capabilities can lead to more interpretable and controllable LLMs.",
  "reply_count": 19,
  "retweet_count": 133,
  "favorite_count": 664,
  "hashtags": [],
  "symbols": [],
  "user_mentions": [],
  "urls": [],
  "media": [
    {
      "media_url": "https://pbs.twimg.com/media/GaLrXtWbsAAmk4X.png",
      "type": "photo"
    }
  ],
  "url": "https://twitter.com/omarsar0/status/1847297594525094081",
  "created_at": "2024-10-18T15:24:02.000Z",
  "#sort_index": "1847297594525094081",
  "view_count": 82155,
  "quote_count": 18,
  "is_quote_tweet": false,
  "is_retweet": false,
  "is_pinned": false,
  "is_truncated": true,
  "startUrl": "https://x.com/omarsar0/status/1847297594525094081"
}