@ZiyuX
Thanks @iScienceLuvr for the summary! Here’s the full thread with more details on the exploration results, training dynamics, and judge calibration: https://t.co/evDxzCvsXw
Viewing enriched Twitter post
Thanks @iScienceLuvr for the summary! Here’s the full thread with more details on the exploration results, training dynamics, and judge calibration: https://t.co/evDxzCvsXw
{
"media": [
{
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2066941356661182762/media_0.jpg",
"media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2066941356661182762/media_0.jpg",
"type": "photo",
"filename": "media_0.jpg"
}
],
"processed_at": "2026-06-16T18:36:37.507104",
"pipeline_version": "2.0"
} {
"type": "tweet",
"id": "2066941356661182762",
"url": "https://x.com/ZiyuX/status/2066941356661182762",
"twitterUrl": "https://twitter.com/ZiyuX/status/2066941356661182762",
"text": "Thanks @iScienceLuvr for the summary! Here’s the full thread with more details on the exploration results, training dynamics, and judge calibration:\n https://t.co/evDxzCvsXw",
"source": "Twitter for iPhone",
"retweetCount": 1,
"replyCount": 0,
"likeCount": 2,
"quoteCount": 0,
"viewCount": 380,
"createdAt": "Tue Jun 16 17:49:55 +0000 2026",
"lang": "en",
"bookmarkCount": 1,
"isReply": false,
"inReplyToId": null,
"conversationId": "2066941356661182762",
"displayTextRange": [
0,
173
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "ZiyuX",
"url": "https://x.com/ZiyuX",
"twitterUrl": "https://twitter.com/ZiyuX",
"id": "394052579",
"name": "Violet X.",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/1505458917312118789/KasJtjd8_normal.jpg",
"coverPicture": "https://pbs.twimg.com/profile_banners/394052579/1460776280",
"description": "",
"location": "United States",
"followers": 191,
"following": 356,
"status": "",
"canDm": false,
"canMediaTag": false,
"createdAt": "Wed Oct 19 14:09:15 +0000 2011",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 822,
"hasCustomTimelines": true,
"isTranslator": false,
"mediaCount": 8,
"statusesCount": 110,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [
"2066933142083207443"
],
"profile_bio": {
"description": "PhD student @Stanford. Working on LLM-based agents",
"entities": {
"description": {
"user_mentions": [
{
"id_str": "",
"indices": [
12,
21
],
"name": "",
"screen_name": "Stanford"
}
]
},
"url": {
"urls": [
{
"display_url": "violetxi.github.io",
"expanded_url": "https://violetxi.github.io",
"indices": [
0,
23
],
"url": "https://t.co/t2zct4pGUX"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {},
"card": null,
"place": {},
"entities": {
"urls": [
{
"display_url": "x.com/ZiyuX/status/2…",
"expanded_url": "https://x.com/ZiyuX/status/2066933142083207443",
"indices": [
150,
173
],
"url": "https://t.co/evDxzCvsXw"
}
],
"user_mentions": [
{
"id_str": "441465751",
"indices": [
7,
20
],
"name": "Tanishq Mathew Abraham, Ph.D.",
"screen_name": "iScienceLuvr"
}
]
},
"quoted_tweet": {
"type": "tweet",
"id": "2066848100447404253",
"url": "https://x.com/iScienceLuvr/status/2066848100447404253",
"twitterUrl": "https://twitter.com/iScienceLuvr/status/2066848100447404253",
"text": "ExpRL: Exploratory RL for LLM Mid-Training\n\nUse RL directly for mid-training. An LLM judge compares the sampled reasoning trace against the reference solution and assigns outcome-level or process-level dense rewards. This lets ExpRL reinforce partial progress, useful intermediate reductions, and productive reasoning behaviors that sparse final-answer rewards often fail to upweight.\n\nOn challenging math reasoning tasks, ExpRL yields stronger RL priming than SFT, sparse-reward GRPO, and self-distillation, and provides a better initialization for subsequent sparse-reward RL.",
"source": "Twitter for iPhone",
"retweetCount": 2,
"replyCount": 1,
"likeCount": 28,
"quoteCount": 0,
"viewCount": 1929,
"createdAt": "Tue Jun 16 11:39:21 +0000 2026",
"lang": "en",
"bookmarkCount": 25,
"isReply": false,
"inReplyToId": null,
"conversationId": "2066848100447404253",
"displayTextRange": [
0,
280
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "iScienceLuvr",
"url": "https://x.com/iScienceLuvr",
"twitterUrl": "https://twitter.com/iScienceLuvr",
"id": "441465751",
"name": "Tanishq Mathew Abraham, Ph.D.",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/2058414948255211520/8jVMerrR_normal.jpg",
"coverPicture": "https://pbs.twimg.com/profile_banners/441465751/1779599917",
"description": "",
"location": "",
"followers": 88471,
"following": 1443,
"status": "",
"canDm": true,
"canMediaTag": true,
"createdAt": "Tue Dec 20 03:45:50 +0000 2011",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 125937,
"hasCustomTimelines": true,
"isTranslator": false,
"mediaCount": 2851,
"statusesCount": 20461,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [
"2066581710847000773"
],
"profile_bio": {
"description": "CEO @SophontAI |\nFounder @MedARC_AI |\nPhD at 19 (2023) |\nex Research Director Stability AI | \nBiomed. engineer @ 14 |\nTEDx talk➡https://t.co/xPxwKTq6Qb",
"entities": {
"description": {
"urls": [
{
"display_url": "bit.ly/3tpAuan",
"expanded_url": "https://bit.ly/3tpAuan",
"indices": [
128,
151
],
"url": "https://t.co/xPxwKTq6Qb"
}
],
"user_mentions": [
{
"id_str": "",
"indices": [
4,
14
],
"name": "",
"screen_name": "SophontAI"
},
{
"id_str": "",
"indices": [
25,
35
],
"name": "",
"screen_name": "MedARC_AI"
}
]
},
"url": {
"urls": [
{
"display_url": "sophont.med",
"expanded_url": "https://sophont.med",
"indices": [
0,
23
],
"url": "https://t.co/MvROZZW1Zg"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {
"media": [
{
"allow_download_status": {
"allow_download": true
},
"display_url": "pic.twitter.com/lkqbbVA4Ua",
"expanded_url": "https://twitter.com/iScienceLuvr/status/2066848100447404253/photo/1",
"ext_master_playlist_only": [],
"ext_media_availability": {
"status": "Available"
},
"ext_playlists": [],
"features": {
"large": {
"faces": []
},
"orig": {
"faces": []
}
},
"id_str": "2066847999922429952",
"indices": [
281,
304
],
"media_key": "3_2066847999922429952",
"media_results": {
"id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARyu604oGmAACgACHK7rZY/bwN0AAA==",
"result": {
"__typename": "ApiMedia",
"id": "QXBpTWVkaWE6DAABCgABHK7rTigaYAAKAAIcrutlj9vA3QAA",
"media_key": "3_2066847999922429952"
}
},
"media_url_https": "https://pbs.twimg.com/media/HK7rTigaYAAQd6O.jpg",
"original_info": {
"focus_rects": [
{
"h": 566,
"w": 1010,
"x": 0,
"y": 0
},
{
"h": 1010,
"w": 1010,
"x": 0,
"y": 0
},
{
"h": 1151,
"w": 1010,
"x": 0,
"y": 0
},
{
"h": 1325,
"w": 663,
"x": 0,
"y": 0
},
{
"h": 1325,
"w": 1010,
"x": 0,
"y": 0
}
],
"height": 1325,
"width": 1010
},
"sizes": {
"large": {
"h": 1325,
"w": 1010
}
},
"type": "photo",
"url": "https://t.co/lkqbbVA4Ua"
}
]
},
"card": null,
"place": {},
"entities": {
"hashtags": [],
"symbols": [],
"urls": [],
"user_mentions": []
},
"quoted_tweet": null,
"retweeted_tweet": null,
"isLimitedReply": false,
"communityInfo": null,
"article": null
},
"retweeted_tweet": null,
"isLimitedReply": false,
"communityInfo": null,
"article": null
}