Billy Xuanming Zhang
@XuanmingZhang07
Interning at GenAI @oracle | PhD student at @ColumbiaCompSci @columbianlp @ColumbiaHCI | Prev at @ETSresearch @TsinghuaNLP | #NLProc #HCI
😵💫 Long-context human-AI planning with LLMs struggles when users have to manually manage all the context in messy chats (e.g. with ChatGPT). Meet 💡JumpStarter: task-structured context curation for better, collaborative planning with LLMs on complex tasks. 🧵 (1/n)
Hi All, we are hiring a postdoc at Columbia CS DAPlab to work on AI agents, blending system+AI. Please send your CV to, [email protected] and Eugene Wu. [email protected] @CUSEAS
Seems like we are not gonna be replaced by AI scientists yet 🥳 It’s been so much fun to contribute one data point to @ChengleiSi’s awesome study. Congrats to the team!!!
Are AI scientists already better than human researchers? We recruited 43 PhD students to spend 3 months executing research ideas proposed by an LLM agent vs human experts. Main finding: LLM ideas result in worse projects than human ideas.
Are AI scientists already better than human researchers? We recruited 43 PhD students to spend 3 months executing research ideas proposed by an LLM agent vs human experts. Main finding: LLM ideas result in worse projects than human ideas.
Killing it!! 👏 Acoustic cues are so important in talker identification! Congrats !!! @ShengyueXiong
🎉Excited to have my first-author paper on talker identification accepted to Interspeech’25! So grateful to my amazing co-author @zcguo81, advisor @bchandra_nu, and everyone at @SoundBrainLab. See you soon in Rotterdam! #Interspeech2025
Check out our new paper, JumpStarter, on how to elicit and manage context interactively in long-session agents.
😵💫 Long-context human-AI planning with LLMs struggles when users have to manually manage all the context in messy chats (e.g. with ChatGPT). Meet 💡JumpStarter: task-structured context curation for better, collaborative planning with LLMs on complex tasks. 🧵 (1/n)
This year, there have been various pieces of evidence that AI agents are starting to be able to conduct scientific research and produce papers end-to-end, at a level where some of these generated papers were already accepted by top-tier conferences/workshops. Intology’s…
🧩 How do we go from examples to structural insight? 🎭 Surface variation in real-world examples makes it hard to spot deeper patterns. ✨ Introducing Schemex: a system for discovering structure through interactive abstraction & contrastive refinement. 📄arxiv.org/pdf/2504.11795
1/🚀Excited to share our AI Realtor project! 🏡 We challenge AI to help sell real estate! This is an interdisciplinary agent work combining economy, personalization & LLM persuasion. More details are below! 👇
Thanks, everyone. I'm very grateful to have gotten a Sloan Fellowship. But I also know how frequent and awful rejection is. Here is my rejected Sloan Research Statement (2023), and my awarded research statement (2025), and the difference between them: lydiachilton.substack.com/p/game-of-pivo…
Assistant Professor of Computer Science Lydia Chilton is one of five Columbia faculty to receive the prestigious Alfred P. Sloan Foundation Research Fellowship! Chilton's work in human-centered AI explores how AI and human creativity can work together to transform design,…
YAY so happy for Lydia @LydiaChilton !!! Well deserved 🎉!!!
🎉Congrats to the 126 early-career scientists who have been awarded a Sloan Research Fellowship this year! These exceptional scholars are drawn from 51 institutions across the US and Canada, and represent the next generation of groundbreaking researchers. sloan.org/fellowships/20…
Exciting post-doc opportunity at the Columbia DAPlab, co-directed by Eugene Wu and myself! We are looking for AI Agents for Fall 2025. 🔥 Check out the details in the job post here: daplab.cs.columbia.edu/positions Please retweet 🙏 hashtag#ColumbiaUnversity hashtag#AI…