🐦 Twitter Post Details

Viewing enriched Twitter post

@hongjin_su

How to adapt LLMs for code 🖥️ to updated libraries and long-tail programming languages w/o training? 🤔 We introduce Arks ⛵️, Active Retrieval in Knowledge Soup, a general pipeline of retrieval-augmented generation for code (RACG). It features: 1️⃣A diverse knowledge soup integrating web search 🌐, documentation 📄, environment feedback 💻, and evolved code snippets 📝; 2️⃣Actively refining queries to retrieve content LLMs prefer; 3️⃣Actively updating knowledge soup with evolved code and environment information ⏩28% 📈in ChatGPT and 23.8% 📈in CodeLLama 👇! Website: https://t.co/xeFzSEK1nI Code: https://t.co/MqpYlU8Lme Paper: https://t.co/9XS0DqB6EG

Media 1
View on Twitter

📊 Media Metadata

{
  "media": [
    {
      "type": "photo",
      "url": "https://pbs.twimg.com/media/GGyvPzxboAAaNGW.jpg",
      "media_url": "https://pbs.twimg.com/media/GGyvPzxboAAaNGW.jpg",
      "filename": "media_0.jpg"
    }
  ],
  "nlp": {
    "sentiment": "positive",
    "processed_at": "2025-08-06T12:53:36.497838"
  },
  "original_structure": "had_media_only",
  "enhanced_from_raw_response": true,
  "enhanced_at": "2025-08-13T17:20:00Z"
}

🔧 Raw API Response

{
  "user": {
    "created_at": "2022-09-07T05:38:52.000Z",
    "default_profile_image": false,
    "description": "Ph.D. student of @HKUniversity, following @taoyds, NLP group 2022. #NLProc",
    "fast_followers_count": 0,
    "favourites_count": 131,
    "followers_count": 304,
    "friends_count": 693,
    "has_custom_timelines": true,
    "is_translator": false,
    "listed_count": 3,
    "location": "Hong Kong",
    "media_count": 8,
    "name": "HongjinSU",
    "normal_followers_count": 304,
    "possibly_sensitive": false,
    "profile_image_url_https": "https://pbs.twimg.com/profile_images/1567388437980008453/CfAW4d17_normal.jpg",
    "screen_name": "hongjin_su",
    "statuses_count": 37,
    "translator_type": "none",
    "url": "https://t.co/T8IvEGVhzK",
    "verified": true,
    "withheld_in_countries": [],
    "id_str": "1567386868295282688"
  },
  "id": "1759978005525643466",
  "conversation_id": "1759978005525643466",
  "full_text": "How to adapt LLMs for code 🖥️ to updated libraries and long-tail programming languages w/o training? 🤔\n\nWe introduce Arks ⛵️, Active Retrieval in Knowledge Soup, a general pipeline of retrieval-augmented generation for code (RACG). It features:\n1️⃣A diverse knowledge soup integrating web search 🌐, documentation 📄, environment feedback 💻, and evolved code snippets 📝;\n2️⃣Actively refining queries to retrieve content LLMs prefer;\n3️⃣Actively updating knowledge soup with evolved code and environment information\n\n⏩28% 📈in ChatGPT and 23.8% 📈in CodeLLama 👇!\n\nWebsite: https://t.co/xeFzSEK1nI\nCode: https://t.co/MqpYlU8Lme\nPaper: https://t.co/9XS0DqB6EG",
  "reply_count": 2,
  "retweet_count": 32,
  "favorite_count": 112,
  "hashtags": [],
  "symbols": [],
  "user_mentions": [],
  "urls": [],
  "media": [
    {
      "media_url": "https://pbs.twimg.com/media/GGyvPzxboAAaNGW.jpg",
      "type": "photo"
    }
  ],
  "url": "https://twitter.com/hongjin_su/status/1759978005525643466",
  "created_at": "2024-02-20T16:27:09.000Z",
  "#sort_index": "1759978005525643466",
  "view_count": 28474,
  "quote_count": 5,
  "is_quote_tweet": false,
  "is_retweet": false,
  "is_pinned": false,
  "is_truncated": true,
  "startUrl": "https://twitter.com/hongjin_su/status/1759978005525643466"
}