🐦 Twitter Post Details

Viewing enriched Twitter post

@TheTuringPost

A new paper from @ylecun and others – V-JEPA 2.1 It changes the recipe of V-JEPA so the model learns both: • Global semantics – what is happening in the scene • Dense spatio-temporal structure – where things are and how they move The idea is to supervise not just masked tokens but the visible ones too There are 4 key ingredients for V-JEPA 2.1: - Dense prediction loss on both masked and visible tokens - Deep self-supervision across intermediate layers - Modality-specific tokenizers (2D for images, 3D for videos) within a shared encoder - Model + data scaling The workflow turns into: masked image/video → encode visible tokens → predict latent representations for both masked and visible tokens → supervise at multiple layers Here are the details:

Media 1

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2034795966931640533/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2034795966931640533/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2026-03-20T12:39:20.076526",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2034795966931640533",
  "url": "https://x.com/TheTuringPost/status/2034795966931640533",
  "twitterUrl": "https://twitter.com/TheTuringPost/status/2034795966931640533",
  "text": "A new paper from @ylecun and others – V-JEPA 2.1\n\nIt changes the recipe of V-JEPA so the model learns both:\n\n• Global semantics – what is happening in the scene\n• Dense spatio-temporal structure – where things are and how they move\n\nThe idea is to supervise not just masked tokens but the visible ones too\n\nThere are 4 key ingredients for V-JEPA 2.1:\n\n- Dense prediction loss on both masked and visible tokens\n- Deep self-supervision across intermediate layers\n- Modality-specific tokenizers (2D for images, 3D for videos) within a shared encoder\n- Model + data scaling\n\nThe workflow turns into: masked image/video → encode visible tokens → predict latent representations for both masked and visible tokens → supervise at multiple layers\n\nHere are the details:",
  "source": "Twitter for iPhone",
  "retweetCount": 27,
  "replyCount": 3,
  "likeCount": 111,
  "quoteCount": 3,
  "viewCount": 14561,
  "createdAt": "Fri Mar 20 00:55:37 +0000 2026",
  "lang": "en",
  "bookmarkCount": 65,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2034795966931640533",
  "displayTextRange": [
    0,
    273
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "TheTuringPost",
    "url": "https://x.com/TheTuringPost",
    "twitterUrl": "https://twitter.com/TheTuringPost",
    "id": "1271482878958940160",
    "name": "Ksenia_TuringPost",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/2019196082727620608/iIXCFAJb_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/1271482878958940160/1677096943",
    "description": "",
    "location": "Join over 102,000 readers",
    "followers": 82849,
    "following": 11420,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Fri Jun 12 16:42:17 +0000 2020",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 13034,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 7355,
    "statusesCount": 18691,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [],
    "profile_bio": {
      "description": "Newsletter exploring AI&ML - AI 101, Agentic Workflow, Business insights. From ML history to AI trends. Led by @kseniase_ Know what you are talking about👇🏼",
      "entities": {
        "description": {
          "hashtags": [],
          "symbols": [],
          "urls": [],
          "user_mentions": [
            {
              "id_str": "0",
              "indices": [
                111,
                121
              ],
              "name": "",
              "screen_name": "kseniase_"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "turingpost.com/subscribe",
              "expanded_url": "https://www.turingpost.com/subscribe",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/81IXvInDft"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/OjHUE73NR4",
        "expanded_url": "https://twitter.com/TheTuringPost/status/2034795966931640533/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": [
              {
                "h": 143,
                "w": 143,
                "x": 260,
                "y": 564
              }
            ]
          },
          "orig": {
            "faces": [
              {
                "h": 143,
                "w": 143,
                "x": 260,
                "y": 564
              }
            ]
          }
        },
        "id_str": "2034795964457000960",
        "indices": [
          274,
          297
        ],
        "media_key": "3_2034795964457000960",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARw9DCYb26AACgACHD0MJq9boNUAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHD0MJhvboAAKAAIcPQwmr1ug1QAA",
            "media_key": "3_2034795964457000960"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HD0MJhvboAARilL.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 770,
              "w": 1375,
              "x": 0,
              "y": 0
            },
            {
              "h": 770,
              "w": 770,
              "x": 0,
              "y": 0
            },
            {
              "h": 770,
              "w": 675,
              "x": 0,
              "y": 0
            },
            {
              "h": 770,
              "w": 385,
              "x": 99,
              "y": 0
            },
            {
              "h": 770,
              "w": 1665,
              "x": 0,
              "y": 0
            }
          ],
          "height": 770,
          "width": 1665
        },
        "sizes": {
          "large": {
            "h": 770,
            "w": 1665
          }
        },
        "type": "photo",
        "url": "https://t.co/OjHUE73NR4"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [],
    "user_mentions": [
      {
        "id_str": "48008938",
        "indices": [
          17,
          24
        ],
        "name": "Yann LeCun",
        "screen_name": "ylecun"
      }
    ]
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "article": null
}