@omarsar0
RT @omarsar0: Interesting paper from Tencent. Tencent trains an agent to manage its own working context, and assigns credit at the level o…
🐦 Twitter Post Details
Viewing enriched Twitter post
📊 Media Metadata
{
"score": 0.38,
"score_components": {
"author": 0.09,
"engagement": 0.0,
"quality": 0.08000000000000002,
"source": 0.135,
"nlp": 0.05,
"recency": 0.025
},
"scored_at": "2026-09-01T09:01:47.167725",
"import_source": "api_import",
"source_tagged_at": "2026-09-01T09:01:47.167738",
"enriched": true,
"enriched_at": "2026-09-01T09:01:47.167740"
}🔧 Raw API Response
{
"type": "tweet",
"id": "2094686810408673402",
"url": "https://x.com/omarsar0/status/2094686810408673402",
"twitterUrl": "https://twitter.com/omarsar0/status/2094686810408673402",
"text": "RT @omarsar0: Interesting paper from Tencent.\n\nTencent trains an agent to manage its own working context, and assigns credit at the level o…",
"source": "Twitter for iPhone",
"retweetCount": 16,
"replyCount": 15,
"likeCount": 101,
"quoteCount": 0,
"viewCount": 9266,
"createdAt": "Tue Sep 01 07:20:27 +0000 2026",
"lang": "en",
"bookmarkCount": 112,
"isReply": false,
"inReplyToId": null,
"conversationId": "2094686810408673402",
"displayTextRange": [
0,
140
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "omarsar0",
"url": "https://x.com/omarsar0",
"twitterUrl": "https://twitter.com/omarsar0",
"id": "3448284313",
"name": "elvis",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/939313677647282181/vZjFWtAn_normal.jpg",
"coverPicture": "https://pbs.twimg.com/profile_banners/3448284313/1565974901",
"description": "",
"location": "DAIR.AI Academy",
"followers": 315733,
"following": 993,
"status": "",
"canDm": true,
"canMediaTag": true,
"createdAt": "Fri Sep 04 12:59:26 +0000 2015",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 39136,
"hasCustomTimelines": true,
"isTranslator": true,
"mediaCount": 4968,
"statusesCount": 19516,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [
"2094499914281566241"
],
"profile_bio": {
"description": "Founder @dair_ai • Prev: Meta AI | PhD • Learn about AI Agents for FREE here: https://t.co/P5SA9u54xO",
"entities": {
"description": {
"urls": [
{
"display_url": "academy.dair.ai/courses/elemen…",
"expanded_url": "https://academy.dair.ai/courses/elements-of-ai-agents",
"indices": [
78,
101
],
"url": "https://t.co/P5SA9u54xO"
}
],
"user_mentions": [
{
"id_str": "",
"indices": [
8,
16
],
"name": "",
"screen_name": "dair_ai"
}
]
},
"url": {
"urls": [
{
"display_url": "dair.ai",
"expanded_url": "https://www.dair.ai/",
"indices": [
0,
23
],
"url": "https://t.co/XQto5ypkSM"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {},
"card": null,
"place": {},
"entities": {
"user_mentions": [
{
"id_str": "3448284313",
"indices": [
3,
12
],
"name": "elvis",
"screen_name": "omarsar0"
}
]
},
"quoted_tweet": null,
"retweeted_tweet": {
"type": "tweet",
"id": "2094505508850032852",
"url": "https://x.com/omarsar0/status/2094505508850032852",
"twitterUrl": "https://twitter.com/omarsar0/status/2094505508850032852",
"text": "Interesting paper from Tencent.\n\nTencent trains an agent to manage its own working context, and assigns credit at the level of individual context edits.\n\nLong-horizon tasks force a model to retrieve, integrate and maintain scattered information across many turns, and keeping every interaction history makes the working context grow without bound.\n\nRecent proactive methods let a model edit its own context with tools, but the toolset stops at search, deletion and summarization.\n\nContextPilot adds global planning, long-term memory and adaptive soft compression, so the agent can offload information rather than only discard it.\n\nThe training side is where it gets interesting. Standard RL hands the final trajectory reward to every intermediate edit equally. ContextPilot uses context and entropy variation to find which editing decisions actually mattered, samples branches at those points, and estimates action-level advantages from all branched trajectories passing through that edit.\n\nOn long-context QA and deep search it beats existing baselines across several base models while holding a more compact working context. Code is available.\n\nPaper: https://t.co/buM6galu4x\n\nChat with Paper: https://t.co/VN5J5PO4Rt",
"source": "Twitter for iPhone",
"retweetCount": 16,
"replyCount": 15,
"likeCount": 101,
"quoteCount": 0,
"viewCount": 9266,
"createdAt": "Mon Aug 31 19:20:01 +0000 2026",
"lang": "en",
"bookmarkCount": 112,
"isReply": false,
"inReplyToId": null,
"conversationId": "2094505508850032852",
"displayTextRange": [
0,
275
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "omarsar0",
"url": "https://x.com/omarsar0",
"twitterUrl": "https://twitter.com/omarsar0",
"id": "3448284313",
"name": "elvis",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/939313677647282181/vZjFWtAn_normal.jpg",
"coverPicture": "https://pbs.twimg.com/profile_banners/3448284313/1565974901",
"description": "",
"location": "DAIR.AI Academy",
"followers": 315733,
"following": 993,
"status": "",
"canDm": true,
"canMediaTag": true,
"createdAt": "Fri Sep 04 12:59:26 +0000 2015",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 39136,
"hasCustomTimelines": true,
"isTranslator": true,
"mediaCount": 4968,
"statusesCount": 19516,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [
"2094499914281566241"
],
"profile_bio": {
"description": "Founder @dair_ai • Prev: Meta AI | PhD • Learn about AI Agents for FREE here: https://t.co/P5SA9u54xO",
"entities": {
"description": {
"urls": [
{
"display_url": "academy.dair.ai/courses/elemen…",
"expanded_url": "https://academy.dair.ai/courses/elements-of-ai-agents",
"indices": [
78,
101
],
"url": "https://t.co/P5SA9u54xO"
}
],
"user_mentions": [
{
"id_str": "",
"indices": [
8,
16
],
"name": "",
"screen_name": "dair_ai"
}
]
},
"url": {
"urls": [
{
"display_url": "dair.ai",
"expanded_url": "https://www.dair.ai/",
"indices": [
0,
23
],
"url": "https://t.co/XQto5ypkSM"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {
"media": [
{
"display_url": "pic.x.com/rEvqA7hzN7",
"expanded_url": "https://x.com/omarsar0/status/2094505508850032852/photo/1",
"ext_master_playlist_only": [],
"ext_media_availability": {
"status": "Available"
},
"ext_playlists": [],
"features": {
"large": {
"faces": [
{
"h": 98,
"w": 98,
"x": 3,
"y": 764
}
]
},
"orig": {
"faces": [
{
"h": 98,
"w": 98,
"x": 3,
"y": 764
}
]
}
},
"id_str": "2094505505490358272",
"indices": [
276,
299
],
"media_key": "3_2094505505490358272",
"media_results": {
"id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAAR0RLaj62lAACgACHREtqcMa4NQAAA==",
"result": {
"__typename": "ApiMedia",
"id": "QXBpTWVkaWE6DAABCgABHREtqPraUAAKAAIdES2pwxrg1AAA",
"media_key": "3_2094505505490358272"
}
},
"media_url_https": "https://pbs.twimg.com/media/HREtqPraUAAEGDe.jpg",
"original_info": {
"focus_rects": [
{
"h": 839,
"w": 1498,
"x": 0,
"y": 0
},
{
"h": 1498,
"w": 1498,
"x": 0,
"y": 0
},
{
"h": 1708,
"w": 1498,
"x": 0,
"y": 0
},
{
"h": 1904,
"w": 952,
"x": 427,
"y": 0
},
{
"h": 1904,
"w": 1498,
"x": 0,
"y": 0
}
],
"height": 1904,
"width": 1498
},
"sizes": {
"large": {
"h": 1904,
"w": 1498
}
},
"type": "photo",
"url": "https://t.co/rEvqA7hzN7"
}
]
},
"card": null,
"place": {},
"entities": {
"hashtags": [],
"symbols": [],
"urls": [
{
"display_url": "arxiv.org/abs/2608.28476",
"expanded_url": "https://arxiv.org/abs/2608.28476",
"indices": [
1154,
1177
],
"url": "https://t.co/buM6galu4x"
},
{
"display_url": "academy.dair.ai/papers/context…",
"expanded_url": "https://academy.dair.ai/papers/contextpilot-teaching-agents-for-proactive-context-management-via-fine-grained-r-2608.28476",
"indices": [
1196,
1219
],
"url": "https://t.co/VN5J5PO4Rt"
}
],
"user_mentions": []
},
"quoted_tweet": null,
"retweeted_tweet": null,
"isLimitedReply": false,
"communityInfo": null,
"article": null
},
"isLimitedReply": false,
"communityInfo": null,
"article": null
}