Assistant professor at POSTECH GSAI/CSE | Prev. UNIST @unc @unccs @uncnlp @AdobeResearch @AmazonScience
Joined September 2009
- Tweets833
- Following522
- Followers410
- Likes665
Hyounghun Kim retweeted
🚨 We are hiring Tenure-Track and Full Professors in important and timely areas such as AI Safety**, Neuroengineering, as part of exciting campus-level strategic cluster hires with CS, and with several positions in the clusters. The Full Professor position in 'Safeguarding AI for Health' is jointly between our strong BioStats department and CS -- please spread the word and feel free to apply & reach out for any questions!
**broad emphasis incl. but not limited to machine learning theory, AI systems, interpretability, robustness, verification, human-AI interaction.
We are hiring at @unccs 🎉
▶️ Two TT Asst. Professors: 1 in Safeguarding AI in Health, 1 in Neuroengineering.
▶️ Full Professor jointly with Biostats.
Part of campus cluster hires (review begins Dec 1).
Plus, Teaching Faculty (review begins Nov 1).
cs.unc.edu/about/faculty-hir…
Hyounghun Kim retweeted
Honored to receive the Outstanding Faculty Mentor Award from the UNC Office of Postdoctoral Affairs, Provost, and Vice Chancellor for Research! 🙏
This one is especially meaningful because mentoring is very close to my heart -- I strongly believe it is one of the main reasons we are in this job, and it is our humble contribution in giving back to the community and helping create the next generation of mentors (students/postdocs/junior faculty, not just in our own lab/department but more broadly across other universities+countries) -- thanks to all my amazing mentees (it's a pleasure and privilege to mentor you) 🤗 💙
Hyounghun Kim retweeted
Excited to share "SPLIT", accepted to #ECCV2026! 🎉
SPLIT is a training-free detector for fully AI-generated and partially edited videos. It targets a deployment-critical setting: high fake recall while falsely flagging only 0.1% of real videos.
Details, paper, and code ↓
Hyounghun Kim retweeted
Deeply happy and honored to be elected as an ACL Fellow -- and to be a part of the respected cohort of this+past years' fellows (congrats everyone)! 🙏
All the credit (and sincere gratitude) to all my amazing students, postdocs, collaborators, mentors, and family! 🤗💙
Hyounghun Kim retweeted
🚨Our paper "Enabling Chatbots with Eyes and Ears" has been accepted to #ACL2025NLP!
👀👂We explore how chatbots can integrate visual, auditory, and textual modalities to support multi-party, multi-session, real-world immersive conversations.
🧵👇
Hyounghun Kim retweeted
Extremely excited to announce that I will be joining @UTAustin @UTCompSci in August 2025 as an Assistant Professor! 🎉
I’m looking forward to continuing to develop AI agents that interact/communicate with people, each other, and the multimodal world. I’ll be recruiting PhD students for Fall 2026 across a range of connected topics (details: esteng.github.io/) and plan on recruiting interns for Fall 2025 as well.
A huge shoutout to my mentors who have supported and shaped my research! Especially grateful to my postdoc advisor @mohitban47 for helping me grow along the whole spectrum of PI skills, and my PhD advisor @ben_vandurme for shaping my trajectory as a researcher, and of course the amazing students/collaborators from @uncnlp and @jhuclsp 🙏
Hyounghun Kim retweeted
In Singapore for #ICLR2025 this week to present the papers + keynotes below 👇, and looking forward to seeing everyone -- happy to chat about research, or faculty+postdoc+phd positions, or simply hanging out (feel free to ping to set up a meeting)! 🙂
Also meet several of our awesome students/postdocs/collaborators who will be presenting this exciting batch of papers (incl. oral+spotlight)!
Hyounghun Kim retweeted
🎉 Congrats to the awesome students, postdocs, & collaborators for this exciting batch of #ICLR2025 and #NAACL2025 accepted papers (FYI some are on the academic/industry job market and a great catch 🙂), on diverse, important topics such as:
-- adaptive data generation environments/policies
-- adapting diverse controls to any diffusion model
-- balancing fast and slow system-1.x planning
-- balancing agents' persuasion resistance+acceptance
-- multimodal compositional+modular video reasoning
-- reverse thinking for stronger LLM reasoning
-- lifelong multimodal instruction tuning via dynamic data selection
-- safe T2I/T2V generation
-- generative infinite games
-- procedural+predictive video representation learning
-- bootstrapping VLN via self-refining data flywheel
-- automated preference data synthesis
-- diagnosing cultural bias of VLMs
-- adaptive decoding to balance contextual+parametric knowledge conflicts
-- positional bias of faithfulness for long-form summarization
-- improving generation faithfulness via multi-agent collaboration
(PS. Also a big thanks to ACs+reviewers for their effort!)
Hyounghun Kim retweeted
Thrilled to announce that Unbounded has been accepted to #ICLR2025 @iclr_conf! 🎉
Experience our character life simulation game, powered entirely by generative models.
We tackle two key challenges:
1️⃣ Ensuring character and environment consistency with a dynamic regional IP-Adapter.
2️⃣ Enabling real-time interaction using a latent consistency model and a distilled specialized LLM.
Check out our work to see the game in action!
See you in Singapore! 🇸🇬
🎉Excited to introduce our new paper: Unbounded: A Generative Game of Character Life Simulation!
We build a game of character life simulation that is fully encapsulated in generative models.
🌟We achieve this with:
▶️ A specialized, distilled LLM that dynamically generates game mechanics, narratives, and character interactions in real-time.
▶️ A dynamic regional IP-Adapter for vision models that ensures consistent yet flexible visual generation of a character across multiple environments.
🧵
I am joining POSTECH after spending a couple of years at UNIST. I really appreciate the strong support from UNIST and am excited for this new journey at POSTECH🚀
Thanks to @mohitban47 for the always-thoughtful advice🙏
Hyounghun Kim retweeted
Complex evaluation is in parts planning and in parts reasoning. Hence, we trained an LLM to think before producing a judgment.
My first work since joining this awesome team 😄
💭🔎 Introducing EvalPlanner – a method to train a Thinking-LLM-as-a-Judge that learns to generate planning & reasoning CoTs for evaluation.
Strong performance on RewardBench, RM-Bench, JudgeBench & FollowBenchEval.
Paper 📄: arxiv.org/abs/2501.18099
Hyounghun Kim retweeted
Many thanks @RealAAAI for selecting me as a #AAAI Fellow! Very humbled+excited to be a part of the respected cohort of this+past years' fellows (and congrats everyone)! 🙏
PS. 100% of the credit goes to all my amazing past & current students+postdocs+collaborators for all their work in these AI areas (& thanks to my mentors+family)! 💙
aaai.org/about-aaai/aaai-awa…
🎉Congratulations to Prof. @mohitban47 on being named a 2025 @RealAAAI Fellow for "significant contributions to multimodal AI foundations & faithful language generation and summarization."👏
16 fellows chosen worldwide by cmte of 9 past fellows & ex-pres: aaai.org/about-aaai/aaai-awa…
Hyounghun Kim retweeted
Deeply honored and humbled to have received the Presidential #PECASE Award by the @WhiteHouse and @POTUS office!
Very grateful to my amazing mentors, students, postdocs, collaborators, and friends+family for making this possible, and for making the journey worthwhile + beautiful 💙 🙏
(Also congrats to all the winners from the last 4-5 years/batches + glad this has been finally announced officially 🙂)
🎉 Congratulations to Prof. @mohitban47 for receiving the Presidential #PECASE Award by @WhiteHouse, which is the highest honor bestowed by US govt. on outstanding scientists/engineers who show exceptional potential for leadership early in their careers!
whitehouse.gov/ostp/news-upd…
ALT Parker Distinguished Professor Mohit Bansal received the Presidential Early Career Award for Scientists and Engineers
Hyounghun Kim retweeted
🚀 New paper out! 📜 Tackling a long-standing challenge in text summarization: How do we properly conduct human evaluations to improve its systematicity, granularity, and scalability? 🤔 Let’s dive in. 🧵👇
📖 Read the full paper here: arxiv.org/abs/2412.07096
Hyounghun Kim retweeted
🚨 I’m on the 2024-2025 academic job market!
j-min.io
I work on ✨ Multimodal AI ✨, with a special focus on enhancing reasoning in both understanding and generation tasks by:
1⃣Making it more scalable
2⃣Making it more faithful
3⃣Evaluating and refining multimodal generation
I’m completing my PhD at @uncnlp with @mohitban47 🙂
I'd love to connect (+ I’ll also be at #NeurIPS2024)
Check the summary of my research: 👇🧵
Hyounghun Kim retweeted
Looking forward to giving this Distinguished Lecture at StonyBrook next week & meeting the several awesome NLP + CV folks there - thanks @b_niranjan + all for the kind invitation 🙂
PS. Excited to give a new talk on "Planning Agents for Collaborative Reasoning and Multimodal Generation" ➡️➡️ diverse, adaptive, and uncertainty-calibrated AI planning agents that can communicate and collaborate for:
(1) robust and grounded multi-agent reasoning and actions (on math, commonsense, coding, games, etc.)
(2) interpretable, controllable planning/programming for multimodal generation (across text, images, videos, audio, layouts, environments).
Detailed abstract in the link below 👇👇👇 (feel free to join)
Join the Dept of #ComputerScience on December 6th at 2:30 PM to welcome Mohit Bansal, UNC Chapel Hill, who will lecture on "Planning Agents for Collaborative Reasoning and Multimodal Generation." Learn more here: tinyurl.com/jkmex3e9
@AI_SBU @CEASSBU @stonybrooku
Hyounghun Kim retweeted
And it's a wrap...signing off now 😴😴
Thanks everyone for joining us at #EMNLP2024 & making it memorable+engaging -- hope y'all enjoyed the program 🤗 ❤️
PS. A big thanks again to @thamar_solorio, @yunnungchen, yaser + all the awesome committee members for all their effort!
Hyounghun Kim retweeted
🚨 How to train LLMs to be persuaded *the right amount* (i.e., not be over-stubborn nor over-gullible)?
Balancing: 1⃣ defending against negative persuasion (makes answers worse / misinfo, jailbreaking, etc.) with 2⃣ accepting positive persuasion (makes answers better)? ➡️➡️➡️ Our multi-agent method, Persuasion-Balanced Training (PBT) recursively creates positive/negative persuasion RLHF data from generated dialogue trees and then trains LLMs to be persuaded when appropriate!
Across 3 models of varying sizes, PBT:
-- improves resistance to misinformation
-- reduces flipflopping
-- obtains best performance on balanced data
-- makes models better teammates
-- improves robustness to ordering sensitivity in multi-agent discussion!
👇👇👇
🚨 Excited to announce “Teaching Models to Balance Resisting and Accepting Persuasion”
LLMs need to be able to reject negative persuasion (e.g. misinfo, jailbreaking, etc)... BUT we argue that they also need to accept positive persuasion (e.g. when they are wrong/unsafe) to be helpful!
We take a 1st step in balancing: 1⃣ defending against negative persuasion (makes answers worse) with 2⃣ accepting positive persuasion (makes answers better), training LLMs to accept persuasion when appropriate.
Our multi-agent method, Persuasion-Balanced Training (PBT) recursively creates RLHF training data from generated dialogue trees and then trains models on positive/negative persuasion.
Across three models of varying sizes, PBT
-- improves resistance to misinformation
-- reduces flipflopping
-- obtains best performance on balanced data
-- makes models better teammates/robust to ordering in multi-agent discussion
When pairing 2 LLMs in a multi-agent debate, we find that pairs generally have a lot of order-dependence. Depending on whether the stronger or weaker model goes first, the team lands on the stronger or weaker answer. PBT reduces this variability and improves team performance.
🧵👇
Hyounghun Kim retweeted
🌴🚨 Join us in Miami (Nov 12-14) for 3 insightful #EMNLP2024 keynotes by @percyliang, @ancadianadragan, and @cocosci_lab on open-source+science, safety+alignment, and Bayes+AI ➡️➡️ 2024.emnlp.org/program/keyno…
PS. Stay tuned -- we are also going to announce an exciting panel soon! 🔥
👇👇👇
🚨 We’re thrilled to announce our exciting keynotes for #EMNLP2024 2024.emnlp.org/program/keyno…
1⃣ 11/12: Percy Liang: open-source+science
2⃣ 11/13: Anca Dragan: safety+alignment
3⃣ 11/14: Tom Griffiths: Bayes+AI
See you in Miami! 🌴
@percyliang @ancadianadragan @cocosci_lab
Multi-session + Multi-party = Mixed-session Conversation🧑🤝🧑
Check out the new #EMNLP2024 Findings paper from @jihyoungjang and @kty4119!
Excited to share that "Mixed-Session Conversation with Egocentric Memory" has been accepted to #EMNLP2024 Findings🎉🎉🎉
Special thanks to @kty4119 and @khsquared for their collaboration!
arxiv.org/abs/2410.02503
