Seungju Han

I’m an incoming Ph.D. student in Computer Science at Stanford and a research intern at NVIDIA advised by Yejin Choi. Before that, I was a visiting student researcher at Allen Institute for AI and Yonsei University, and worked at Hyperconnect, which was acquired by Match Group for $1.7B. I received my B.S. in ECE from Seoul National University in 2024.

I work on improving foundational models (such as LLMs and multimodal LLMs), and especially interested in data, training algorithms, and evaluation. I prefer very simple and clean ideas that are scalable. These days, I think a lot about reasoning and knowledge of LLMs.

Research (Show all)

Nemotron-CrossThink: Scaling Self-Learning beyond Math Reasoning
Syeda Nahida Akter, Shrimai Prabhumoye, Matvei Novikov, Seungju Han, Ying Lin, Evelina Bakhturi, Eric Nyberg, Yejin Choi, Mostofa Patwary, Mohammad Shoeybi, Bryan Catanzaro
Preprint, 2025
Paper

Prismatic Synthesis & G-Vendi Score: How Data Diversification makes R1-32B a Better Teacher than R1-671B
Jaehun Jung, Seungju Han*, Ximing Lu*, Skyler Hallinan*, Shrimai Prabhumoye, Mostafa Patwary, Mohammad Shoeybi, Bryan Catanzaro, Yejin Choi
Blogpost, 2025
Blog

Retro-Search: Exploring Untaken Paths for Deeper and Efficient Reasoning
Ximing Lu*, Seungju Han*, David Acuna*, Hyunwoo Kim*, Jaehun Jung*, Shrimai Prabhumoye, Niklas Muennighoff, Mostofa Patwary, Mohammad Shoeybi, Bryan Catanzaro, Yejin Choi
Preprint, 2025
We retrospectively search for efficient reasoning paths using long-thought reasoning LLMs, and train with these paths leads to better performance and inference-time efficiency.
Paper

Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models
NVIDIA Team, contributed to build pretraining data
Tech report, 2025
Paper

Representation Bending for Large Language Model Safety
Ashkan Yousefpour*, Taeheon Kim*, Ryan S. Kwon, Seungbeen Lee, Wonje Jeung, Seungju Han, Alvin Wan, Harrison Ngan, Youngjae Yu, Jonghyun Choi
ACL 2025
Paper

MAPoRL: Multi-Agent Post-Co-Training for Collaborative Large Language Models with Reinforcement Learning
Chanwoo Park, Seungju Han, Xingzhi Guo, Asuman Ozdaglar, Kaiqing Zhang, Joo-Kyung Kim
ACL 2025
We introduce a multi-agent RL to teach LLMs to do collaboration, which can't be achieved with SFT.
Paper

AI as Humanity's Salieri: Quantifying Linguistic Creativity of Language Models via Systematic Attribution of Machine Text against Web Text
Ximing Lu, Melanie Sclar, Skyler Hallinan, Niloofar Mireshghallah, Jiacheng Liu, Seungju Han, Allyson Ettinger, Liwei Jiang, Khyathi Chandu, Nouha Dziri, Yejin Choi
ICLR 2025 (Oral, 1.8% acceptance rate)
We introduce Creativity Index to quantify how LM outputs are linguistically creative when compared to the web corpus.
Paper / Code

Do LLMs Have Distinct and Consistent Personality? TRAIT: Personality Testset designed for LLMs with Psychometrics
Seungbeen Lee*, Seungwon Lim*, Seungju Han, Giyoung Oh, Minju Kim, Beongwoo Kwak, Jiwan Chung, Hyungjoo Chae, Dongha Lee, Jinyoung Yeo, Youngjae Yu
NAACL Findings 2025
We developed a testbed, TRAIT, to assess the personality of LLMs with psychometrics.
Paper

WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs
Seungju Han*, Kavel Rao*, Allyson Ettinger, Liwei Jiang, Bill Yuchen Lin, Nathan Lambert, Yejin Choi, Nouha Dziri
NeurIPS 2024
We built a state-of-the-art, open-sourced moderation model for LLMs.
Paper / Code / Data

WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models
Liwei Jiang, Kavel Rao*, Seungju Han*, Allyson Ettinger, Faeze Brahman, Sachin Kumar, Niloofar Mireshghallah, Ximing Lu, Maarten Sap, Yejin Choi, Nouha Dziri
NeurIPS 2024
We developed a framework that uses real-world user-LLM interactions to discover 5.7K unique jailbreak tactics (10x larger than prior resources) that reveal vulnerabilities of LLMs.
Paper / Code / Data

Selective Vision is the Challenge for Visual Reasoning: A Benchmark for Visual Argument Understanding
Jiwan Chung, Sungje Lee, Minseo Kim, Seungju Han, Ashkan Yousefpour, Jack Hessel, Youngjae Yu
EMNLP 2024 Oral
Paper

Multimodal Laughter Reasoning with Textual Audio-Visual Representation
Hyun Lee, Sung Bin Kim, Seungju Han, Youngjae Yu, Tae Hyun Oh
NAACL 2024
Paper

Reading Books is Great, But Not if You Are Driving! Visually Grounded Reasoning about Defeasible Commonsense Norms
Seungju Han, Junhyeok Kim, Jack Hessel, Liwei Jiang, Jiwan Chung, Yejin Son, Yejin Choi, Youngjae Yu
EMNLP 2023 Oral
We introduced commonsense reasoning task that require visual grounding and released high-quality evaluation datasets for these tasks, setting a foundation to narrow the gap between humans and models.
Paper

CHAMPAGNE: Learning Real-world Conversation from Large-Scale Web Videos
Seungju Han, Jack Hessel, Nouha Dziri, Yejin Choi, Youngjae Yu
ICCV 2023
Using internet-scale web videos, we created a visual-grounded dialogue dataset with 18M items (20x larger than previous resources) to train vision-language models.
Paper / Code

Measuring and Improving Semantic Diversity of Dialogue Generation
Seungju Han, Beomsu Kim, Buru Chang
EMNLP 2022
We measured the diversity of responses generated by LMs using the semantic space with embeddings.
Paper

Meet Your Favorite Character: Open-domain Chatbot Mimicking Fictional Characters with only a Few Utterances
Seungju Han*, Beomsu Kim*, Jin Yong Yoo*, Seokjun seo, Sangbum Kim, Enkhbayar Erdenee, Buru Chang
NAACL 2022
We introduced an in-context learning algorithm to generate responses in the style of fictional characters with only a few utterances.
Paper

Understanding and Improving the Exemplar-based Generation for Open-domain Conversation
Seungju Han*, Beomsu Kim*, Seokjun Seo*, Enkhbayar Erdenee*, Buru Chang
ACL 2022 4th Workshop on NLP4ConvAI Oral Presentation, Outstanding Paper
Developed an algorithm to successfully train a retrieval-augmented conversational model.
Paper

Distilling the Knowledge of Large-scale Generative Models into Retrieval Models for Efficient Open-domain Conversation
Beomsu Kim*, Seokjun Seo*, Seungju Han*, Enkhbayar Erdenee*, Buru Chang
EMNLP 2021
We designed an algorithm for knowledge distillation from generative LMs to retrieval conversational models, achieving speedups of 20–40 times compared to the teacher model while maintaining performance.
Paper

Disentangling Label Distribution for Long-tailed Visual Recognition
Youngkyu Hong*, Seungju Han*, Kwanghee Choi*, Seokjun Seo, Beomsu Kim, Buru Chang
CVPR 2021
We designed the new training objective to correct biases in output probabilities when training on long-tail distributions.
Paper / Code

Attentron: Few-Shot Text-to-Speech Utilizing Attention-Based Variable-Length Embedding
Seungwoo Choi*, Seungju Han*, Dongyoung Kim*, Sungjoo Ha
Interspeech 2020
Paper