Hui Chen

I packaged up the "autoresearch" project into a new self-contained minimal repo if people would like to play over the weekend. It's basically nanochat LLM training core stripped down to a single-GPU, one file version of ~630 lines of code, then: - the human iterates on the prompt (.md) - the AI agent iterates on the training code (.py) The goal is to engineer your agents to make the fastest research progress indefinitely and without any of your own involvement. In the image, every dot is a complete LLM training run that lasts exactly 5 minutes. The agent works in an autonomous loop on a git feature branch and accumulates git commits to the training script as it finds better settings (of lower validation loss by the end) of the neural network architecture, the optimizer, all the hyperparameters, etc. You can imagine comparing the research progress of different prompts, different agents, etc. github.com/karpathy/autor… Part code, part sci-fi, and a pinch of psychosis :)

The boundary between reality and the virtual world is about to disappear. Imagine characters from Pokemon or Digimon coming to life—not on a screen, but right in front of you. You reach out, and they respond. Introducing X1. The first real-time interactive video model that brings imagination into reality. Built by the Xmax AI team from Tsinghua University. Let’s Play the World through AI. Repost & comment to join the X1 private beta (free).

@allen_ai @NlpWestlake @henryzhao4321 @ChengleiSi @May_F1_ Lastly, Prof. Min-Yen Kan @knmnyn shared some new thoughts on AI research, where the research manifold is now being disrupted by the new entrant of artificially intelligent agents and tools.

@allen_ai @NlpWestlake @henryzhao4321 @ChengleiSi Prof. Yi R. (May) Fung @May_F1_ presented several great research projects on building autonomous AI scientists.

Thrilled to have successfully hosted the AAAI 2026 Workshop on AI for Scientific Research in Singapore🇸🇬! Sincere thanks to all speakers, authors, and participants for the contributions!👏 Hope this workshop inspires more collaborations and ideas on AI for scientific research!

🧑‍💻For those interested, the slides for all the keynote speakers are uploaded and shared onto the 𝗔𝗔𝗔𝗜'𝟮𝟲 𝗔𝗜 𝗳𝗼𝗿 𝗦𝗰𝗶𝗲𝗻𝘁𝗶𝗳𝗶𝗰 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗪𝗼𝗿𝗸𝘀𝗵𝗼𝗽 webpage publicly at ai4research-workshop.github.io 🌟𝗣𝗲𝘁𝗲𝗿 𝗖𝗹𝗮𝗿𝗸: 𝗙𝗿𝗼𝗺 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗧𝗼𝗼𝗹𝘀 𝘁𝗼 𝗟𝗼𝗻𝗴-𝗛𝗼𝗿𝗶𝘇𝗼𝗻 𝗔𝗜-𝗔𝘀𝘀𝗶𝘀𝘁𝗲𝗱 𝗦𝗰𝗶𝗲𝗻𝗰𝗲 🌟𝗬𝘂𝗲 𝗭𝗵𝗮𝗻𝗴: 𝗔𝗜-𝗱𝗿𝗶𝘃𝗲𝗻 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗼𝗻 🌟𝗖𝗵𝗲𝗻 𝗭𝗵𝗮𝗼: 𝗔𝗜 𝗳𝗼𝗿 𝗦𝗰𝗶𝗲𝗻𝘁𝗶𝗳𝗶𝗰 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝘄𝗶𝘁𝗵 𝗕𝗲𝘁𝘁𝗲𝗿 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗦𝗲𝗮𝗿𝗰𝗵 𝗦𝘆𝘀𝘁𝗲𝗺𝘀 🌟𝗖𝗵𝗲𝗻𝗴𝗹𝗲𝗶 𝗦𝗶: 𝗧𝗼𝘄𝗮𝗿𝗱𝘀 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗲𝗱 𝗔𝗜 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 🌟𝗬𝗶 𝗥. (𝗠𝗮𝘆) 𝗙𝘂𝗻𝗴: 𝗕𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝘁𝗵𝗲 𝗔𝘂𝘁𝗼𝗻𝗼𝗺𝗼𝘂𝘀 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗦𝗰𝗶𝗲𝗻𝘁𝗶𝘀𝘁, 𝗙𝗿𝗼𝗺 𝗘𝘅𝗽𝗹𝗼𝗿𝗮𝘁𝗶𝗼𝗻 𝗠𝗲𝗰𝗵𝗮𝗻𝗶𝘀𝗺𝘀 𝘁𝗼 𝗦𝗰𝗵𝗼𝗹𝗮𝗿𝗹𝘆 𝗜𝗻𝘁𝗲𝗴𝗿𝗶𝘁𝘆 🌟𝗠𝗶𝗻-𝗬𝗲𝗻 𝗞𝗮𝗻: 𝗧𝗵𝗲 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗠𝗮𝗻𝗶𝗳𝗼𝗹𝗱

🤩Great line-up of speakers at the AAAI'26 AI for Scientific Research workshop today! Thrilled to share our latest contributions toward this important direction as well. In particular, check out: 1️⃣𝙒𝙚𝙗𝙬𝙖𝙩𝙘𝙝𝙚𝙧: 𝘽𝙧𝙚𝙖𝙠𝙞𝙣𝙜 𝙣𝙚𝙬 𝙛𝙧𝙤𝙣𝙩𝙞𝙚𝙧 𝙤𝙛 𝙫𝙞𝙨𝙞𝙤𝙣-𝙡𝙖𝙣𝙜𝙪𝙖𝙜𝙚 𝙙𝙚𝙚𝙥 𝙧𝙚𝙨𝙚𝙖𝙧𝙘𝙝 𝙖𝙜𝙚𝙣𝙩 2️⃣𝘾𝙞𝙩𝙚𝙂𝙪𝙖𝙧𝙙: 𝙁𝙖𝙞𝙩𝙝𝙛𝙪𝙡 𝘾𝙞𝙩𝙖𝙩𝙞𝙤𝙣 𝘼𝙩𝙩𝙧𝙞𝙗𝙪𝙩𝙞𝙤𝙣 𝙛𝙤𝙧 𝙇𝙇𝙈𝙨 𝙫𝙞𝙖 𝙍𝙚𝙩𝙧𝙞𝙚𝙫𝙖𝙡-𝘼𝙪𝙜𝙢𝙚𝙣𝙩𝙚𝙙 𝙑𝙖𝙡𝙞𝙙𝙖𝙩𝙞𝙤𝙣 3️⃣𝙑𝙚𝙧𝙞-𝙍1: 𝙏𝙤𝙬𝙖𝙧𝙙 𝙋𝙧𝙚𝙘𝙞𝙨𝙚 𝙖𝙣𝙙 𝙁𝙖𝙞𝙩𝙝𝙛𝙪𝙡 𝘾𝙡𝙖𝙞𝙢 𝙑𝙚𝙧𝙞𝙛𝙞𝙘𝙖𝙩𝙞𝙤𝙣 𝙫𝙞𝙖 𝙊𝙣𝙡𝙞𝙣𝙚 𝙍𝙚𝙞𝙣𝙛𝙤𝙧𝙘𝙚𝙢𝙚𝙣𝙩 𝙇𝙚𝙖𝙧𝙣𝙞𝙣𝙜 4️⃣𝙀𝙭𝙥𝙡𝙤𝙧𝙞𝙣𝙜 𝘼𝙜𝙚𝙣𝙩𝙞𝙘 𝙈𝙪𝙡𝙩𝙞𝙢𝙤𝙙𝙖𝙡 𝙇𝙖𝙧𝙜𝙚 𝙇𝙖𝙣𝙜𝙪𝙖𝙜𝙚 𝙈𝙤𝙙𝙚𝙡𝙨: 𝘼 𝙎𝙪𝙧𝙫𝙚𝙮 𝙛𝙤𝙧 𝘼𝙄𝙎𝙘𝙞𝙚𝙣𝙩𝙞𝙨𝙩𝙨 And more, from our collaborators and team! 🎉🎉

📢 We are excited to release the schedule for AI4Research workshop at #AAAI2026! Join us at 📍Garnet Room 214, Singapore EXPO on Jan 26. We are delighted to host an incredible lineup of invited speakers! @henryzhao4321 @ChengleiSi @May_F1_ @knmnyn #AI4Science #ScientificDiscovery

Our poster session will be at 11am - 2pm, Dec 5. Please feel free to drop by our poster at Exhibit Hall C,D,E #2305!

I’ll be at #NeurIPS2025 next week presenting MLR-Bench! Reach out if you want to chat: AI4Science, coding agents, self-evolving etc.

@ManasiSharma_ @scale_AI Great work!

🚀New @scale_AI paper: 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵𝗥𝘂𝗯𝗿𝗶𝗰𝘀, a benchmark for evaluating Deep Research (DR) agents. Even top agents like Gemini & OpenAI DR achieve <𝟲𝟴% 𝗿𝘂𝗯𝗿𝗶𝗰 𝗰𝗼𝗺𝗽𝗹𝗶𝗮𝗻𝗰𝗲. We built 𝟮.𝟱𝗞+ expert rubrics with 𝟮.𝟴𝗞+ hrs of human labor to measure why.

📢We have extended the submission deadline to November 15, 2025 (AoE)!

📢CFP: Join us for the "AI for Scientific Research" workshop at AAAI 2026 in Singapore (Jan 26, 2026)! ai4research-workshop.github.io 📅We invite submissions of papers by Oct 22, 2025: openreview.net/group?id=AAAI.… #AAAI2026 #AI4Research

TTD-DR is a Deep Researcher agent that models research report writing as a diffusion process, where a first draft is gradually polished into a high-quality final version. Read about its impressive results in long-form report writing and complex reasoning tasks: goo.gle/4ncW5yp