{
  "type": "tweet",
  "id": "2072380547603829224",
  "url": "https://x.com/LiorOnAI/status/2072380547603829224",
  "twitterUrl": "https://twitter.com/LiorOnAI/status/2072380547603829224",
  "text": "Most world models predict what happens next.\n\nSora predicts pixels, JEPA compresses observations. \n\nNEO tries to figure out why something happened instead.\n\nExample: show it a shape moving left then down, and instead of just reconstructing that motion, it learns \"left\" and \"down\" as separate reusable building blocks then reuses them elsewhere.\n\nInstead of one big black-box model, NEO searches for a short \"program\" made of simple reusable steps that explains what it saw. \n\nThe interesting bit isn't that it learns programs. It's that it discovers the building blocks of explanation on its own, no labels, no hand-coded symbols, just raw observation.",
  "source": "Twitter for iPhone",
  "retweetCount": 5,
  "replyCount": 3,
  "likeCount": 23,
  "quoteCount": 0,
  "viewCount": 4052,
  "createdAt": "Wed Jul 01 18:03:20 +0000 2026",
  "lang": "en",
  "bookmarkCount": 12,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2072380547603829224",
  "displayTextRange": [
    0,
    280
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "LiorOnAI",
    "url": "https://x.com/LiorOnAI",
    "twitterUrl": "https://twitter.com/LiorOnAI",
    "id": "931470139",
    "name": "Lior Alexander",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/2032256308196564993/ozddLZ2O_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/931470139/1761077189",
    "description": "",
    "location": "San Francisco, CA",
    "followers": 115949,
    "following": 2391,
    "status": "",
    "canDm": true,
    "canMediaTag": false,
    "createdAt": "Wed Nov 07 07:19:36 +0000 2012",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 7002,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 665,
    "statusesCount": 3813,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [],
    "profile_bio": {
      "description": "Founder @AlphaSignalAI • MIT Lecturer • ex-MILA researcher • In AI for the last 10 years.",
      "entities": {
        "description": {
          "user_mentions": [
            {
              "id_str": "",
              "indices": [
                8,
                22
              ],
              "name": "",
              "screen_name": "AlphaSignalAI"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "alphasignal.ai",
              "expanded_url": "https://alphasignal.ai",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/AyubevaLcb"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {},
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [],
    "user_mentions": []
  },
  "quoted_tweet": {
    "type": "tweet",
    "id": "2072230094614237320",
    "url": "https://x.com/SungjinAhn_/status/2072230094614237320",
    "twitterUrl": "https://twitter.com/SungjinAhn_/status/2072230094614237320",
    "text": "🚀 We introduce Neural Theorizer (NEO) — a new type of world model that learns to theorize the world from observation, without language or LLM supervision.\n\nSelected as an ICML 2026 oral presentation — 0.7% of submitted papers.\n\nThe paper asks:\n\n\"What does it mean to understand the world and build a world model?\"\n\nToday’s world models are often trained to predict the future: the next frame, next latent state, or next observation.\n\nBut is prediction enough?\n\nWe argue that a world model should be a theory-building system: one that discovers reusable primitives, composes them into executable explanations, and transfers those explanations to novel phenomena.\n\nNEO is our first step toward this vision — a World Theory Model that learns explicit, compositional theories from raw observation.\n\nThis work was led by my wonderful students: Doojin Baek*(@doojin_a_baek), Gyubin Lee* (@gyubin0521), Junyeob Baek (@JunyeobB), and Hosung Lee (@HosungLee_).\n\nFor more details, take a look at the paper — and if you’re attending ICML, let’s talk there!\n\n📄 arXiv: https://t.co/TGMXLLfzP7\n🌐 Project page: https://t.co/aLJywp8rfq",
    "source": "Twitter for iPhone",
    "retweetCount": 84,
    "replyCount": 8,
    "likeCount": 494,
    "quoteCount": 6,
    "viewCount": 30100,
    "createdAt": "Wed Jul 01 08:05:29 +0000 2026",
    "lang": "en",
    "bookmarkCount": 480,
    "isReply": false,
    "inReplyToId": null,
    "conversationId": "2072230094614237320",
    "displayTextRange": [
      0,
      277
    ],
    "inReplyToUserId": null,
    "inReplyToUsername": null,
    "author": {
      "type": "user",
      "userName": "SungjinAhn_",
      "url": "https://x.com/SungjinAhn_",
      "twitterUrl": "https://twitter.com/SungjinAhn_",
      "id": "48712353",
      "name": "Sungjin Ahn",
      "isVerified": false,
      "isBlueVerified": true,
      "verifiedType": null,
      "profilePicture": "https://pbs.twimg.com/profile_images/1979851432284000256/DjVeQi64_normal.jpg",
      "coverPicture": "https://pbs.twimg.com/profile_banners/48712353/1643629367",
      "description": "",
      "location": "Daejeon, Korea",
      "followers": 3231,
      "following": 1327,
      "status": "",
      "canDm": false,
      "canMediaTag": true,
      "createdAt": "Fri Jun 19 14:00:30 +0000 2009",
      "entities": {
        "description": {
          "urls": []
        },
        "url": {}
      },
      "fastFollowersCount": 0,
      "favouritesCount": 2757,
      "hasCustomTimelines": true,
      "isTranslator": false,
      "mediaCount": 128,
      "statusesCount": 625,
      "withheldInCountries": [],
      "affiliatesHighlightedLabel": {},
      "possiblySensitive": false,
      "pinnedTweetIds": [
        "2072230094614237320"
      ],
      "profile_bio": {
        "description": "Prof@KAIST, Chief Dreamer of Machine Learning & Mind Lab https://t.co/ato9yodtm5",
        "entities": {
          "description": {
            "urls": [
              {
                "display_url": "mlml.kaist.ac.kr",
                "expanded_url": "http://mlml.kaist.ac.kr",
                "indices": [
                  57,
                  80
                ],
                "url": "https://t.co/ato9yodtm5"
              }
            ]
          },
          "url": {
            "urls": [
              {
                "display_url": "sungjinahn.com",
                "expanded_url": "http://www.sungjinahn.com",
                "indices": [
                  0,
                  23
                ],
                "url": "https://t.co/h63vg1dwIh"
              }
            ]
          }
        }
      },
      "isAutomated": false,
      "automatedBy": null
    },
    "extendedEntities": {
      "media": [
        {
          "additional_media_info": {
            "monetizable": false
          },
          "display_url": "pic.twitter.com/TLw8YR2VsK",
          "expanded_url": "https://twitter.com/SungjinAhn_/status/2072230094614237320/video/1",
          "ext_master_playlist_only": [],
          "ext_media_availability": {
            "status": "Available"
          },
          "ext_playlists": [],
          "id_str": "2072227536084611072",
          "indices": [
            278,
            301
          ],
          "media_key": "13_2072227536084611072",
          "media_results": {
            "id": "QXBpTWVkaWFSZXN1bHRzOgwABAoAARzCB/cGGzAAAAA=",
            "result": {
              "__typename": "ApiMedia",
              "id": "QXBpTWVkaWE6DAAECgABHMIH9wYbMAAAAA==",
              "media_key": "13_2072227536084611072"
            }
          },
          "media_url_https": "https://pbs.twimg.com/amplify_video_thumb/2072227536084611072/img/rjKyJ54ZFg-Y7BaS.jpg",
          "original_info": {
            "focus_rects": [],
            "height": 808,
            "width": 2160
          },
          "sizes": {
            "large": {
              "h": 766,
              "w": 2048
            }
          },
          "type": "video",
          "url": "https://t.co/TLw8YR2VsK",
          "video_info": {
            "aspect_ratio": [
              270,
              101
            ],
            "duration_millis": 27621,
            "variants": [
              {
                "content_type": "application/x-mpegURL",
                "url": "https://video.twimg.com/amplify_video/2072227536084611072/pl/5Il4Uz1Re3nb3g03.m3u8?tag=28"
              },
              {
                "bitrate": 256000,
                "content_type": "video/mp4",
                "url": "https://video.twimg.com/amplify_video/2072227536084611072/vid/avc1/720x270/W8rFtEU-qhHf__dn.mp4?tag=28"
              },
              {
                "bitrate": 832000,
                "content_type": "video/mp4",
                "url": "https://video.twimg.com/amplify_video/2072227536084611072/vid/avc1/962x360/Eb9RjquYapsNENGn.mp4?tag=28"
              },
              {
                "bitrate": 2176000,
                "content_type": "video/mp4",
                "url": "https://video.twimg.com/amplify_video/2072227536084611072/vid/avc1/1924x720/TDAaUOiXCcj2eEJ_.mp4?tag=28"
              },
              {
                "bitrate": 10368000,
                "content_type": "video/mp4",
                "url": "https://video.twimg.com/amplify_video/2072227536084611072/vid/avc1/2160x808/awuMfEnYLy0Nvn2D.mp4?tag=28"
              }
            ]
          }
        },
        {
          "display_url": "pic.twitter.com/TLw8YR2VsK",
          "expanded_url": "https://twitter.com/SungjinAhn_/status/2072230094614237320/video/1",
          "ext_master_playlist_only": [],
          "ext_media_availability": {
            "status": "Available"
          },
          "ext_playlists": [],
          "id_str": "2072227585166372864",
          "indices": [
            278,
            301
          ],
          "media_key": "16_2072227585166372864",
          "media_results": {
            "id": "QXBpTWVkaWFSZXN1bHRzOgwAAgoAARzCCAJzm3AACgACHMIKSrpbEIgAAA==",
            "result": {
              "__typename": "ApiMedia",
              "id": "QXBpTWVkaWE6DAACCgABHMIIAnObcAAKAAIcwgpKulsQiAAA",
              "media_key": "16_2072227585166372864"
            }
          },
          "media_url_https": "https://pbs.twimg.com/tweet_video_thumb/HMIIAnObcAADEUB.jpg",
          "original_info": {
            "focus_rects": [],
            "height": 832,
            "width": 1882
          },
          "sizes": {
            "large": {
              "h": 832,
              "w": 1882
            }
          },
          "type": "animated_gif",
          "url": "https://t.co/TLw8YR2VsK",
          "video_info": {
            "aspect_ratio": [
              941,
              416
            ],
            "variants": [
              {
                "bitrate": 0,
                "content_type": "video/mp4",
                "url": "https://video.twimg.com/tweet_video/HMIIAnObcAADEUB.mp4"
              }
            ]
          }
        }
      ]
    },
    "card": null,
    "place": {},
    "entities": {
      "hashtags": [],
      "symbols": [],
      "urls": [
        {
          "display_url": "arxiv.org/abs/2605.03413",
          "expanded_url": "https://arxiv.org/abs/2605.03413",
          "indices": [
            1056,
            1079
          ],
          "url": "https://t.co/TGMXLLfzP7"
        },
        {
          "display_url": "doojinbaek.github.io/publications/l…",
          "expanded_url": "https://doojinbaek.github.io/publications/learning-to-theorize-the-world/",
          "indices": [
            1096,
            1119
          ],
          "url": "https://t.co/aLJywp8rfq"
        }
      ],
      "user_mentions": [
        {
          "id_str": "1835286498457116672",
          "indices": [
            852,
            866
          ],
          "name": "Doojin Baek",
          "screen_name": "doojin_a_baek"
        },
        {
          "id_str": "1515988297755279360",
          "indices": [
            882,
            893
          ],
          "name": "GyuBin Lee",
          "screen_name": "gyubin0521"
        },
        {
          "id_str": "1539170459933380608",
          "indices": [
            910,
            919
          ],
          "name": "Junyeob Baek @ ICML26 🇰🇷",
          "screen_name": "JunyeobB"
        },
        {
          "id_str": "2072228168577212416",
          "indices": [
            938,
            949
          ],
          "name": "Hosung Lee",
          "screen_name": "HosungLee_"
        }
      ]
    },
    "quoted_tweet": null,
    "retweeted_tweet": null,
    "isLimitedReply": false,
    "communityInfo": null,
    "article": null
  },
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}