🐦 Twitter Post Details

Viewing enriched Twitter post

@dahou_yasser

We are releasing Falcon Perception, an open-vocabulary referring expression segmentation model. Along with it, a 0.3B OCR model that is on par with 3-10x larger competitors. Current systems solve this with complex pipelines (separate encoders, late fusion, matching algorithms). We developed a novel simpler "bitter" approach: one early-fusion Transformer (image + text from first layer) with a shared parameter space, and let scale + training signal do the work. Please check our work ! 📄 Paper: https://t.co/dWvK5t7MIt 💻 Code: https://t.co/AJ65GbMrUY 🎮 Playground: https://t.co/BIgisZkeid 🤗 Blogpost: https://t.co/J2IjlBPywF

Media 2
Media 3

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039242378809385331/media_0.mp4",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039242378809385331/media_0.mp4",
      "type": "video",
      "filename": "media_0.mp4"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039242378809385331/media_1.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039242378809385331/media_1.jpg",
      "type": "photo",
      "filename": "media_1.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039242378809385331/media_2.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039242378809385331/media_2.jpg",
      "type": "photo",
      "filename": "media_2.jpg"
    }
  ],
  "processed_at": "2026-04-01T10:51:36.469371",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2039242378809385331",
  "url": "https://x.com/dahou_yasser/status/2039242378809385331",
  "twitterUrl": "https://twitter.com/dahou_yasser/status/2039242378809385331",
  "text": "We are releasing Falcon Perception, an open-vocabulary referring expression segmentation model. Along with it, a 0.3B OCR model that is on par with  3-10x larger competitors.\n\nCurrent systems solve this with complex pipelines (separate encoders, late fusion, matching algorithms). We developed a novel simpler \"bitter\" approach: one early-fusion Transformer (image + text from first layer) with a shared parameter space, and let scale + training signal do the work. Please check our work !\n\n📄 Paper: https://t.co/dWvK5t7MIt\n💻 Code: https://t.co/AJ65GbMrUY\n🎮 Playground: https://t.co/BIgisZkeid\n🤗 Blogpost: https://t.co/J2IjlBPywF",
  "source": "Twitter for iPhone",
  "retweetCount": 31,
  "replyCount": 5,
  "likeCount": 116,
  "quoteCount": 4,
  "viewCount": 7226,
  "createdAt": "Wed Apr 01 07:24:05 +0000 2026",
  "lang": "en",
  "bookmarkCount": 82,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2039242378809385331",
  "displayTextRange": [
    0,
    280
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "dahou_yasser",
    "url": "https://x.com/dahou_yasser",
    "twitterUrl": "https://twitter.com/dahou_yasser",
    "id": "1073197020041658368",
    "name": "Yasser Dahou",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1985652404403261440/hn50K1qZ_normal.jpg",
    "coverPicture": "",
    "description": "",
    "location": "Abu Dhabi, UAE",
    "followers": 243,
    "following": 748,
    "status": "",
    "canDm": true,
    "canMediaTag": false,
    "createdAt": "Thu Dec 13 12:44:43 +0000 2018",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 1169,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 48,
    "statusesCount": 592,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [],
    "profile_bio": {
      "description": "Researcher (Technology Innovation Institute, PhD at Dublin City University)",
      "entities": {
        "description": {
          "hashtags": [],
          "symbols": [],
          "urls": [],
          "user_mentions": []
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "additional_media_info": {
          "monetizable": true
        },
        "display_url": "pic.twitter.com/1RZuPtuZAE",
        "expanded_url": "https://twitter.com/dahou_yasser/status/2039242378809385331/video/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "id_str": "2039242226669428736",
        "indices": [
          281,
          304
        ],
        "media_key": "13_2039242226669428736",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwABAoAARxM2AAzm6AAAAA=",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAAECgABHEzYADOboAAAAA==",
            "media_key": "13_2039242226669428736"
          }
        },
        "media_url_https": "https://pbs.twimg.com/amplify_video_thumb/2039242226669428736/img/jX7WRO5a1nnE9r6N.jpg",
        "original_info": {
          "focus_rects": [],
          "height": 1080,
          "width": 1920
        },
        "sizes": {
          "large": {
            "h": 1080,
            "w": 1920
          }
        },
        "type": "video",
        "url": "https://t.co/1RZuPtuZAE",
        "video_info": {
          "aspect_ratio": [
            16,
            9
          ],
          "duration_millis": 70983,
          "variants": [
            {
              "content_type": "application/x-mpegURL",
              "url": "https://video.twimg.com/amplify_video/2039242226669428736/pl/5yC-gLglC6-5AdCd.m3u8?tag=21&v=cfc"
            },
            {
              "bitrate": 256000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2039242226669428736/vid/avc1/480x270/JDYP0ik8Yp7zqwoy.mp4?tag=21"
            },
            {
              "bitrate": 832000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2039242226669428736/vid/avc1/640x360/6HemHq6i-3HZg5uM.mp4?tag=21"
            },
            {
              "bitrate": 2176000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2039242226669428736/vid/avc1/1280x720/ws8prALYMM8McY8y.mp4?tag=21"
            },
            {
              "bitrate": 10368000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2039242226669428736/vid/avc1/1920x1080/Gb-BWfORBeQ9lJd5.mp4?tag=21"
            }
          ]
        }
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "timestamps": [],
    "urls": [
      {
        "display_url": "arxiv.org/pdf/2603.27365",
        "expanded_url": "https://arxiv.org/pdf/2603.27365",
        "indices": [
          500,
          523
        ],
        "url": "https://t.co/dWvK5t7MIt"
      },
      {
        "display_url": "github.com/tiiuae/falcon-…",
        "expanded_url": "https://github.com/tiiuae/falcon-perception",
        "indices": [
          532,
          555
        ],
        "url": "https://t.co/AJ65GbMrUY"
      },
      {
        "display_url": "vision.falcon.aidrc.tii.ae",
        "expanded_url": "https://vision.falcon.aidrc.tii.ae",
        "indices": [
          570,
          593
        ],
        "url": "https://t.co/BIgisZkeid"
      },
      {
        "display_url": "huggingface.co/blog/tiiuae/fa…",
        "expanded_url": "https://huggingface.co/blog/tiiuae/falcon-perception/",
        "indices": [
          606,
          629
        ],
        "url": "https://t.co/J2IjlBPywF"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "article": null
}