Defended my thesis yesterday :) Its been a fantastic ride at Columbia NLP and I am grateful to my advisor Smaranda Muresan for believing in my work. Special thanks to Violet Peng who introduced me to Creative NLG which made a lot of the work in my thesis possible

GREAT NEWS: we have the amazing Tiffany Tseng 🍡 joining soon :) so excited!!! now Columbia Human-Computer Interaction Group is becoming bigger and bigger for sure ✨

🌟 Delighted to introduce our #DIS24 work: PodReels - an AI-powered Premiere Pro extension that supports video teaser creation. Check out how PodReels helps tackle this creative task 🎙️🎞️ and the insights we gained from developing AI-powered extensions! 🤖🔌

another great mind joining! we are so so so excited :) now, we have HCI x AI for health! — Orson is also looking for PhDs and Postdocs in the coming cycle! Please apply and join Columbia Human-Computer Interaction Group 🦁💙

Hello English second-language learners, have you ever struggled with writing coherent English essays in your language exams? ✨Introducing DECOR: Improving Coherence in L2 English Writing with a Novel Benchmark for Incoherence Detection, Reasoning, and Rewriting 🧵(1/n)

Please follow @Tweetorial_Lab🐦‍⬛ we are a group of students working with @lydiachilton at Columbia Human-Computer Interaction Group trying to talk about 🧪science & technology 💻 thru your life experiences! 🧬🔭🔬👩‍🔬 fun & engaging! ⬇️🛒🛍️ see how we talk about your amazon impulsive buying experience :)

✨ Storytelling is at the heart of many nonprofits, but few have the resources to consistently transform those stories into compelling content to inspire their audiences to action. Many get left untold. How can we help amplify these messages? Meet MoodSmith🕊️featured in #ICCC24

🎏#CHI2025 authors! looking for longitudinal evidence for your AI tool? our #DIS2024 work show AI tools remain (& become more!!) useful over time. plus, we explore novelty effect, learning curve, performance, customization & appropriation in 3 week doi.org/10.1145/364383… (1/n)

Small distraction from the #CHI2025 deadline with Columbia Human-Computer Interaction Group friends at Din Tai Fung

New paper on human-AI interaction. We hire 18 writers to edit quirks in AI writing & see if #AI can mimic this process to improve its own writing Verdict: Writer-edited > AI-edited > AI-generated In other words:🚨Edits enhance alignment in writing🚨 🔗arxiv.org/pdf/2409.14509

✈️ Heading to Pittsburgh tomorrow for #UIST2024! Excited to present "StreetNav," where we explored using street cameras to support precise outdoor navigation for blind pedestrians👩‍🦯 Center for Smart Streetscapes Brian A. Smith Gil Zussman

🤳 The #UIST2024 Social Media Challenge kicks off tomorrow! Can’t wait to see all the fun and creative things you share!!

always looking for speakers👀 interested? [email protected]

Had a lot of fun visiting Columbia Human-Computer Interaction Group last week! Cross country flight with 5 month old, not so much 😅

What does the rise of #ArtificialIntelligence mean for the future of #creativity? 🎨 In this article, I describe the relationship between mechanics & artistry, exploring how history can guide our creative future. 🔗aiandstorytelling.substack.com/p/mechanics-an…… #aiart #artist #creativetech #hci

Super excited to share that my first, first-author paper has been accepted to #CHI2025​​ ​🌸🇯🇵! Our system, DynEx bridges design and engineering to help users flesh out their ideas through a Design Matrix and generate code to create a functional app 🖥️ 📖: arxiv.org/abs/2410.00400

🧩 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

updated my site in time #CHI2025 🇯🇵🗼 not presenting, but happy to chat about: 🦆 FeedQUAC: ambient creativity tool with real-time audio feedback (with autodesk) 🎎 DoubleAgents: AI agents that learn human values/policies by simulating test cases & co-planning, executing them!!

😵‍💫 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)

what if AI played the role of ur sassy gay bestie 🏳️‍🌈 or AAVE-speaking friend 👋🏾 ? you: plan a trip? 🤖 AI: “Yasss queen! let’s werk this babe✨💅” LLMs can talk like us, but it shapes how we trust, rely on & relate to them…🧵 📣 our #FAccT2025 paper: bit.ly/3HJ6rWI