🐦 Twitter Post Details

Viewing enriched Twitter post

@dylanjcastillo

Structured outputs can decrease LLM's performance in some tasks. I replicated @willkurt / @dottxtai rebuttal of Let Me Speak Freely? (LMSF) using gpt-4o-mini. The rebuttal correctly highlights many flaws with the original study, but ironically, LMSF's conclusion still holds. https://t.co/m3UTIsziA2

Media 1

šŸ“Š Media Metadata

{
  "links_checked": true,
  "checked_at": "2025-08-10T10:33:08.040290",
  "media": [
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1867154938759155750/media_0.jpg?",
      "filename": "media_0.jpg"
    }
  ],
  "reprocessed_at": "2025-08-12T15:27:54.852390",
  "reprocessed_reason": "missing_media_array",
  "original_structure": "had_media_only"
}

šŸ”§ Raw API Response

{
  "user": {
    "created_at": "2020-03-20T22:27:50.000Z",
    "default_profile_image": false,
    "description": "šŸ™‹ā€ā™‚ļø About: https://t.co/zEVJvdkpb3\nšŸ›  Side projects: https://t.co/jpXZeVmfqN",
    "fast_followers_count": 0,
    "favourites_count": 3449,
    "followers_count": 467,
    "friends_count": 426,
    "has_custom_timelines": true,
    "is_translator": false,
    "listed_count": 10,
    "location": "Madrid, Spain",
    "media_count": 150,
    "name": "Dylan",
    "normal_followers_count": 467,
    "possibly_sensitive": false,
    "profile_banner_url": "https://pbs.twimg.com/profile_banners/1241129207553556482/1585064539",
    "profile_image_url_https": "https://pbs.twimg.com/profile_images/1858923622187851776/6Xtnp6Pa_normal.jpg",
    "screen_name": "dylanjcastillo",
    "statuses_count": 899,
    "translator_type": "none",
    "url": "https://t.co/qBzLhoeVTF",
    "verified": false,
    "withheld_in_countries": [],
    "id_str": "1241129207553556482"
  },
  "id": "1867154938759155750",
  "conversation_id": "1867154938759155750",
  "full_text": "Structured outputs can decrease LLM's performance in some tasks.\n\nI replicated @willkurt / @dottxtai rebuttal of Let Me Speak Freely? (LMSF) using gpt-4o-mini.\n\nThe rebuttal correctly highlights many flaws with the original study, but ironically, LMSF's conclusion still holds. https://t.co/m3UTIsziA2",
  "reply_count": 4,
  "retweet_count": 2,
  "favorite_count": 29,
  "hashtags": [],
  "symbols": [],
  "user_mentions": [
    {
      "id_str": "5011881",
      "name": "Will Kurt",
      "screen_name": "willkurt",
      "profile": "https://twitter.com/willkurt"
    },
    {
      "id_str": "1719934067435520000",
      "name": ".txt",
      "screen_name": "dottxtai",
      "profile": "https://twitter.com/dottxtai"
    }
  ],
  "urls": [],
  "media": [
    {
      "media_url": "https://pbs.twimg.com/media/Gel3hXrbwAA1nvl.jpg",
      "type": "photo"
    }
  ],
  "url": "https://twitter.com/dylanjcastillo/status/1867154938759155750",
  "created_at": "2024-12-12T10:30:02.000Z",
  "#sort_index": "1867154938759155750",
  "view_count": 6101,
  "quote_count": 2,
  "is_quote_tweet": false,
  "is_retweet": false,
  "is_pinned": false,
  "is_truncated": false,
  "startUrl": "https://x.com/dylanjcastillo/status/1867154938759155750"
}