Juhao Liang

Excited to share our work: Alignment at Pre-training! Towards Native Alignment for Arabic LLMs 🎉 📄 Published at NeurIPS ‘24 📍 NeurIPS East Exhibit Hall A-C #1803, Vancouver 🗓️ Fri 13 Dec 16:30 - 19:30 arxiv: arxiv.org/abs/2412.03253 poster: neurips.cc/virtual/2024/p…

Efficiently Democratizing Medical LLMs for 50 Languages via a Mixture of Language Family Experts Datasets: - huggingface.co/datasets/Freed… (Likes: 0) - huggingface.co/datasets/Freed… (Likes: 0) Models: - huggingface.co/models/Freedom… (Likes: 2, Downloads: 33) - huggingface.co/models/Freedom… (Likes: 2, Downloads: 33) - huggingface.co/models/Freedom… (Likes: 2, Downloads: 23) - huggingface.co/models/Freedom… (Likes: 1, Downloads: 32) - huggingface.co/models/Freedom… (Likes: 0, Downloads: 37) - huggingface.co/models/Freedom… (Likes: 0, Downloads: 71) - huggingface.co/models/Freedom… (Likes: 0, Downloads: 18) - huggingface.co/models/nicobos… (Likes: 0, Downloads: 0) Tags: [safetensors, biology, medical, question-answering, multi-language, license:apache-2.0]

𝗦𝗺𝘂𝗿𝗳𝘀 𝗮𝗿𝗲 𝗿𝗲𝗮𝗹! 𝗮𝘁𝗹𝗲𝗮𝘀𝘁 𝗶𝗻 𝗔𝗜! A new research paper called "Smurfs: Leveraging Multiple Proficiency Agents with Context-Efficiency for Tool Planning" proposes a framework that transforms 𝗟𝗟𝗠𝘀 𝗶𝗻𝘁𝗼 𝗮 𝘀𝘆𝗻𝗲𝗿𝗴𝗶𝘀𝘁𝗶𝗰 𝗲𝗻𝘀𝗲𝗺𝗯𝗹𝗲.

Phoenix: Democratizing ChatGPT across Languages Releases a LLM ‘Phoenix’, achieving competitive performance on English and Chinese while excelling in low-resources languages. repo: github.com/FreedomIntelli… abs: arxiv.org/abs/2304.10453

Listening to @dyaaa from @SignalHQ presenting in-person @GlasgowCS his thoughts and perspectives entitled Applied Research for Product Development” #IRTalk

Data scientists: "My back hurts so bad" Also data scientists, while their models train: