Viewing enriched Twitter post
Humans can see in high-res, high-FPS in real-time. Why can't VLMs? Introducing AutoGaze: ViTs/VLMs "gaze" only at key video regions! Up to 4-100x token savings, 19x speedup, and enables scaling to 4K-res 1K-frame videos. 📄 https://t.co/GhbWZwMAg7 🌐 https://t.co/mEJ991MAIR 🤗 https://t.co/FOfc2QRThi (1/n)🧵
{
"media": [
{
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036514959228805269/media_0.mp4",
"media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036514959228805269/media_0.mp4",
"type": "video",
"filename": "media_0.mp4"
},
{
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036514959228805269/media_2.jpg",
"media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036514959228805269/media_2.jpg",
"type": "photo",
"filename": "media_2.jpg"
}
],
"processed_at": "2026-04-03T15:51:25.395756",
"pipeline_version": "2.0"
}{
"type": "tweet",
"id": "2036514959228805269",
"url": "https://x.com/baifeng_shi/status/2036514959228805269",
"twitterUrl": "https://twitter.com/baifeng_shi/status/2036514959228805269",
"text": "Humans can see in high-res, high-FPS in real-time. Why can't VLMs?\n\nIntroducing AutoGaze: ViTs/VLMs \"gaze\" only at key video regions! Up to 4-100x token savings, 19x speedup, and enables scaling to 4K-res 1K-frame videos.\n\n📄 https://t.co/GhbWZwMAg7\n🌐 https://t.co/mEJ991MAIR\n🤗 https://t.co/FOfc2QRThi\n\n(1/n)🧵",
"source": "Twitter for iPhone",
"retweetCount": 198,
"replyCount": 47,
"likeCount": 1563,
"quoteCount": 22,
"viewCount": 146308,
"createdAt": "Tue Mar 24 18:46:17 +0000 2026",
"lang": "en",
"bookmarkCount": 1052,
"isReply": false,
"inReplyToId": null,
"conversationId": "2036514959228805269",
"displayTextRange": [
0,
276
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "baifeng_shi",
"url": "https://x.com/baifeng_shi",
"twitterUrl": "https://twitter.com/baifeng_shi",
"id": "1443398093374824452",
"name": "Baifeng",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/1747514937587707904/PA__jsSx_normal.jpg",
"coverPicture": "https://pbs.twimg.com/profile_banners/1443398093374824452/1633073074",
"description": "",
"location": "",
"followers": 2246,
"following": 692,
"status": "",
"canDm": false,
"canMediaTag": true,
"createdAt": "Thu Sep 30 02:11:42 +0000 2021",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 1368,
"hasCustomTimelines": true,
"isTranslator": false,
"mediaCount": 53,
"statusesCount": 402,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [
"2036514959228805269"
],
"profile_bio": {
"description": "@physical_int, @berkeley_ai, vision, robotics; ex @nvidia; opinions are my own.",
"entities": {
"description": {
"hashtags": [],
"symbols": [],
"urls": [],
"user_mentions": [
{
"id_str": "0",
"indices": [
0,
13
],
"name": "",
"screen_name": "physical_int"
},
{
"id_str": "0",
"indices": [
15,
27
],
"name": "",
"screen_name": "berkeley_ai"
},
{
"id_str": "0",
"indices": [
50,
57
],
"name": "",
"screen_name": "nvidia"
}
]
},
"url": {
"urls": [
{
"display_url": "bfshi.github.io",
"expanded_url": "http://bfshi.github.io",
"indices": [
0,
23
],
"url": "https://t.co/qYlJcBWAcd"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {
"media": [
{
"additional_media_info": {
"monetizable": true
},
"allow_download_status": {
"allow_download": true
},
"display_url": "pic.twitter.com/O0A0WLrkxb",
"expanded_url": "https://twitter.com/baifeng_shi/status/2036514959228805269/video/1",
"ext_media_availability": {
"status": "Available"
},
"id_str": "2036514530247999488",
"indices": [
277,
300
],
"media_key": "13_2036514530247999488",
"media_results": {
"id": "QXBpTWVkaWFSZXN1bHRzOgwABAoAARxDJyzjWoAAAAA=",
"result": {
"__typename": "ApiMedia",
"id": "QXBpTWVkaWE6DAAECgABHEMnLONagAAAAA==",
"media_key": "13_2036514530247999488"
}
},
"media_url_https": "https://pbs.twimg.com/amplify_video_thumb/2036514530247999488/img/mO7F83AA3Kvmzkjg.jpg",
"original_info": {
"focus_rects": [],
"height": 1080,
"width": 1920
},
"sizes": {
"large": {
"h": 1080,
"w": 1920
}
},
"type": "video",
"url": "https://t.co/O0A0WLrkxb",
"video_info": {
"aspect_ratio": [
16,
9
],
"duration_millis": 94677,
"variants": [
{
"content_type": "application/x-mpegURL",
"url": "https://video.twimg.com/amplify_video/2036514530247999488/pl/0c6PXXoCW4d4LMnA.m3u8?tag=21&v=cfc"
},
{
"bitrate": 256000,
"content_type": "video/mp4",
"url": "https://video.twimg.com/amplify_video/2036514530247999488/vid/avc1/480x270/UBtMdRIxt2760FlE.mp4?tag=21"
},
{
"bitrate": 832000,
"content_type": "video/mp4",
"url": "https://video.twimg.com/amplify_video/2036514530247999488/vid/avc1/640x360/9D0ikVMF58253Frh.mp4?tag=21"
},
{
"bitrate": 2176000,
"content_type": "video/mp4",
"url": "https://video.twimg.com/amplify_video/2036514530247999488/vid/avc1/1280x720/aMx2LpozmFrTMU3w.mp4?tag=21"
},
{
"bitrate": 10368000,
"content_type": "video/mp4",
"url": "https://video.twimg.com/amplify_video/2036514530247999488/vid/avc1/1920x1080/W5nM7sfw9JRk7c2f.mp4?tag=21"
}
]
}
}
]
},
"card": null,
"place": {},
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [
{
"display_url": "arxiv.org/abs/2603.12254",
"expanded_url": "http://arxiv.org/abs/2603.12254",
"indices": [
225,
248
],
"url": "https://t.co/GhbWZwMAg7"
},
{
"display_url": "autogaze.github.io",
"expanded_url": "http://autogaze.github.io",
"indices": [
251,
274
],
"url": "https://t.co/mEJ991MAIR"
},
{
"display_url": "huggingface.co/collections/bf…",
"expanded_url": "http://huggingface.co/collections/bfshi/autogaze",
"indices": [
277,
300
],
"url": "https://t.co/FOfc2QRThi"
}
],
"user_mentions": []
},
"quoted_tweet": null,
"retweeted_tweet": null,
"isLimitedReply": false,
"article": null
}