Kihoon Son
@kihoon_son
Ph.D. Student @kixlab_kaist. I am highly interested in HAI approaches for capturing and utilizing human tacit knowledge in a designer's workflow.
Design has long been considered mysterious, driven by intuition and creativity🧐 But we set out to demystify it! Our paper investigates designers' tacit knowledge and proposes guidelines for systems that can capture and share it This work got an Honorable Mention 🏆 #chi2024 🧵
Im so back. I gonna present CodeShaping tomorrow 10am in G304 at #CHI2025 !! Ill try to do a live demo, see you there!
I'm in Yokohama for #CHI2025 to present my work #Proxona! 😊 I'll be presenting at 9:12 AM in Room G314+G315. If you're interested in user simulation & prototyping, data-driven personas, conversational interactions, or supporting sensemaking and ideation, please come!
Exciting news 🪩 I'll present our upcoming paper Proxona: Supporting Creators' Sensemaking and Ideation with LLM-Powered Audience Personas at #CHI2025 in Yokohama! 🇯🇵 Can we simulate virtual yet plausible and meaningful audience personas for content creators?
Evaluating language model responses on open-ended tasks is hard! 🤔 We introduce EvalAgent, a framework that identifies nuanced and diverse criteria 📋✍️. EvalAgent identifies 👩🏫🎓 expert advice on the web that implicitly address the user’s prompt 🧵👇
Huge congratulations to my coauthors @farwest1138 @jsquare & LiYi on receiving the @acm_dis best paper award 🏆 for Narrative Motion Blocks. This paper is very close to my heart, as it combines direct manipulation affordances with textual prompts for animation creation.
My flight just diverted back to Lisbon from Funchal coz the 🌬️, and there is no more flight available to change until the end of #DIS2025 conference… where I supposed to present To Search or To Gen Well, I’ll just chill in Lisbon then
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.
The #DynamicAbstractions Reading Group returns this Friday! 🎉 📍Topic: Simulations and Understanding 🗓️ Date & Time: June 27, 12pm EST / 9am PST 🎙️ Speaker: Andy Matuschak (andymatuschak.org) ✉️ Full letter: buttondown.com/dynamic_abstra… 🔁 Please #RT to share! 🧵 More:
안녕하세요, @kixlab_kaist 에서 만든 📢 공약쏙쏙 시스템📢을 소개합니다! (promisescope.com) 🤔공약쏙쏙? #대선2025 공약을 바탕으로 공약이 가져올 수 있는 사회적 영향, 그리고 '나'에게 어떤 변화가 생길 수 있을지 LLM (GPT-4o) 을 활용해 자동으로 보여주는 시스템입니다. (1/3)
🌸 Thanks everyone for coming! 🌸#CHI2025 #DynamicAbstractions
🌸 #CHI2025 meetup! At CHI'24, we had an informal meetup for folks into dynamic abstractions and more. Since then: UIST'24 workshop, reading group, and Discord chats. This year, we'll have small-group lunches on Mon 4/28. Add your name by Sun 04/27: shorturl.at/QZ3Cw 🧵
We're at LBW-167 again now! #CHI2025 (Mandatory selfie again!
We will also present INSPO (arxiv.org/abs/2311.16521) poster today at #CHI2025. Come say hi!
Excited to be at #CHI2025! In our paper "Paratrouper," we explore how AI-driven image-generation techniques can empower creatives to explore a variety of visual design possibilities for casts of characters. 💖🐸🦜👩🚀🧙♀️🎭💖 @ADSKResearch @UofT @medialab @FluidInterfaces 🧵(1/6)
Building more cooperative and user-centered LLM interactions? 🤝 Inspired by Gricean Maxims Theory, we derived 9 design insights across the HAI interaction cycle! ✨ Check out our LBW poster at #CHI2025! 📄 dl.acm.org/doi/10.1145/37… 📍 programs.sigchi.org/chi/2025/progr…
🔎 How can we make LLM interactions more cooperative and human-like? We applied Gricean Maxims — principles for good conversation — to human-LLM interaction and derived 9 design insights! Come see our LBW poster on Wed at #CHI2025! Let’s talk on #HAI, #AIAlignment, #UserIntent 😆
This is happening tomorrow at 2:10 PM at the Coding and Development session at G418-419! Come by and check out how a proactive agent with its own cursor editing the code with you impacts AI programming usage and developer experience!
🤖👩🏻💻Proactive AI tools like @GitHubCopilot @cursor_ai @allhands_ai promise to assist developers by anticipating their needs and automating engineering processes—but do they truly help? We evaluated three design probes to explore the trade-offs of proactive AI programming support
🎨 Generative models enhance the graphic design process, but limit broad exploration. We built Expandora: structured input + LLM-based prompt generation + mindmap UI to boost creativity! 🚀 Come to our LBW poster at #CHI2025🌸 📄dl.acm.org/doi/10.1145/37… 📍programs.sigchi.org/chi/2025/progr…
Are you interested in expanding your design ideas with generative AI? 🎨🚀 Come check out our LBW poster on Expandora at #CHI2025 on Wed. We explored how structured input and a mindmap-style interface can enable broader exploration. Love to chat about #GenAI #Creativity #Design
I'll be presenting "Understanding the LLM-ification of CHI" **Wed, 30 Apr, 2:10 PM - 3:40 PM** at 📍G401📍 Come to the #CHI2025 paper session "Future of HCI and Research Practices" for all the wonderful papers! Let's connect and chat in Japan!! 😀
🤔Interested in how #HCI thinks about using #LLMs, or looking to understand best practices for human-LLM interaction? 🚨🚨New paper: Understanding the LLM-ification of CHI: Unpacking the Impact of LLMs at CHI through a Systematic Literature Review
What if AI didn't wait for developer questions, but could anticipate their needs? Our new #CHI2025 paper rethinks designing AI assistants for programming—from reactive tools to proactive partners.
🤖LLMs are being integrated everywhere, but how do we ensure they're delivering meaningful user experiences? In our #chi2025 paper, we empower designers to think about this via 🎨designerly adaptation🎨 of LLMs and built a Figma widget to help! 📜 arxiv.org/abs/2401.09051 🧵👇
One of our first papers, led by Juhoon Lee @juhoonjulie, to be presented at #CHI2025, visits a classic context for online safety: online MOBA games. This paper challenges the narrative that conflict can be prevented by facilitating smoother communication. joseph.seering.org/papers/Lee_et_…
🤖👩🏻💻Proactive AI tools like @GitHubCopilot @cursor_ai @allhands_ai promise to assist developers by anticipating their needs and automating engineering processes—but do they truly help? We evaluated three design probes to explore the trade-offs of proactive AI programming support
I've been thinking a lot about the interaction pattern with agents recently. I really enjoyed reading @cursor_ai 's blog posts on problems below: cursor.com/blog/problems-… cursor.com/blog/problems-… Three key ideas stood out. 1. The focus on context, especially multi-hop context…