Seungtaek Choi

News

• Aug 2025: Two papers got accepted in CIKM 2025.
• Jun 2025: One paper got accepted in TACL and will be presented at EMNLP 2025.
• Apr 2024: EEVE-Korean has been awarded as the "LLM of the Year" (Pretrained) based on the Open Ko-LLM Leaderboard.
• Oct 2023: I got ranked at 1st in Open LLM Leaderboard! (model name: Riiid/sheep-duck-llama-2-70b-v1.1)

Publications J C W A

• C25Towards Fully-Automated Materials Discovery via Large-Scale Synthesis Dataset and Expert-Level LLM-as-a-Judge [PDF]
  Heegyu Kim, Taeyang Jeon, Seungtaek Choi, Ji Hoon Hong, Dong Won Jeon, Ga-Yeon Baek, Gyeong-Won Kwak, Dong-Hee Lee, Jisu Bae, Chihoon Lee, Yunseo Kim, Seon-Jin Choi, Jin-Seong Park, Sung Beom Cho, Hyunsouk Cho^†
  CIKM 2025
  
• C24Rethinking the Training Paradigm of Discrete Token-Based Multimodal LLMs: An Analysis of Text-Centric Bias
  Wansik Jo, Jooyeong Na, Soyeon Hong, Seungtaek Choi, Hyunsouk Cho^†
  CIKM 2025
  
• J4Overcoming Source Object Grounding for Semantic Image Editing
  Yeonjoon Jung, Seungtaek Choi, Seung-won Hwang^†
  TACL 2025
  
• A2Trillion 7B Technical Report [PDF]
  Sungjun Han, Juyoung Suk, Suyeong An, Hyungguk Kim, Kyuseok Kim, Wonsuk Yang, Seungtaek Choi, Jamin Shin (Trillion Labs)
  arXiv 2025 (technical report of Trillion-7B-preview)
  
• C23FLEX: Expert-level False-Less EXecution Metric for Reliable Text-to-SQL Benchmark [PDF]
  Heegyu Kim, Taeyang Jeong, Seunghwan Choi, Seungtaek Choi, Hyunsouk Cho^†
  NAACL 2025
  
• J3ScoreCL: Augmentation-Adaptive Contrastive Learning via Score-Matching Function [PDF]
  Jin-Young Kim^†, Soonwoo Kwon, Hyojun Go, Yunsung Lee, Seungtaek Choi, Hyun-Gyoon Kim^†
  Machine Learning 2025
  
• C22Interventional Speech Noise Injection for ASR Generalizable Spoken Language Understanding [PDF]
  Yeonjoon Jung, Jaseseong Lee, Seungtaek Choi, Dohyeon Lee, Minsoo Kim, Seung-won Hwang^†
  EMNLP 2024
  
• A1Efficient and Effective Vocabulary Expansion Towards Multilingual Large Language Models [PDF]
  Seungduk Kim*, Seungtaek Choi*, Myeongho Jeong (* equal contribution)
  arXiv 2024 (technical report of EEVE-Korean)
  
• C21Multi-Architecture Multi-Expert Diffusion Models [PDF]
  Yunsung Lee*, JinYoung Kim*, Hyojun Go*, Myeongho Jeong, Shinhyeok Oh, Seungtaek Choi^† (* equal contribution)
  AAAI 2024
  
• C20Addressing Negative Transfer in Diffusion Models [PDF]
  Hyojun Go*, JinYoung Kim*, Yunsung Lee*, Seunghyun Lee*, Shinhyeok Oh, Hyeongdon Moon, Seungtaek Choi^† (* equal contribution)
  NeurIPS 2023
  
• C19Addressing Cold Start Problem for End-to-end Automatic Speech Scoring [PDF]
  Jungbae Park, Seungtaek Choi^†
  INTERSPEECH 2023
  
• C18Cross Encoding As Augmentation: Towards Effective Educational Text Classification [PDF]
  Hyun Seung Lee*, Seungtaek Choi*^†, Yunsung Lee, Hyeongdon Moon, Shinhyeok Oh, Myeongho Jeong, Hyojun Go, Christian Wallraven (* equal contribution)
  Findings of ACL 2023
  
• C17Evaluation of Question Generation Needs More References [PDF]
  Shinhyeok Oh*, Hyojun Go*, Hyeongdon Moon, Yunsung Lee, Myeongho Jeong, Hyun Seung Lee, Seungtaek Choi^† (* equal contribution)
  Findings of ACL 2023
  
• C16Retrieval-augmented Instructional Video Encoding for Dense Video Captioning [PDF]
  Yeonjoon Jung, Seungtaek Choi, Seung-won Hwang^†, Jihyuk Kim, Minji Seo, Minsoo Kim
  Findings of ACL 2023
  
• C15On Complementarity Objectives for Hybrid Retrieval [PDF]
  Dohyeon Lee, Seung-won Hwang^†, Kyungjae Lee, Seungtaek Choi, Sunghyun Park
  ACL 2023
  
• C14Towards Practical Plug-and-Play Diffusion Models [PDF]
  Hyojun Go*, Yunsung Lee*, JinYoung Kim*, Seunghyun Lee, Myeongho Jeong, Hyun Seung Lee, Seungtaek Choi^† (* equal contribution)
  CVPR 2023
  
• C13Evaluating the Knowledge Dependency of Questions [PDF]
  Hyeongdon Moon*, Yoonseok Yang*, Jamin Shin, Hangyeol Yu, Seunghyun Lee, Myeongho Jeong, Juneyoung Park, Minsam Kim, Seungtaek Choi^† (* equal contribution)
  EMNLP 2022
  
• C12Towards Compositional Generalization in Code Search [PDF]
  Hojae Han, Seung-won Hwang^†, Shuai Lu, Nan Duan, Seungtaek Choi
  EMNLP 2022 (short)
  
• C11Debiasing Event Understanding for Visual Commonsense Tasks [PDF]
  Minji Seo*, Yeonjoon Jung*, Seungtaek Choi, Seung-won Hwang^†, Bei Liu (* equal contribution)
  Findings of ACL 2022
  
• C10C²L: Causally Contrastive Learning for Robust Text Classification [PDF]
  Seungtaek Choi*, Myeongho Jeong*, Hojae Han, Seung-won Hwang^† (* equal contribution)
  AAAI 2022
  
• C9Structure-Augmented Keyphrase Generation [PDF]
  Jihyuk Kim, Myeongho Jeong, Seungtaek Choi, Seung-won Hwang^†
  EMNLP 2021
  
• C8Counterfactual Generative Smoothing for Imbalanced Natural Language Classification [PDF]
  Hojae Han, Seungtaek Choi, Myeongho Jeong, Jin-woo Park, Seung-won Hwang^†
  CIKM 2021 (short)
  
• J2Label and Context Augmentation for Response Selection at DSTC8 [PDF]
  Myeongho Jeong*, Seungtaek Choi*, Jinyoung Yeo, Seung-won Hwang^† (* equal contribution)
  TASLP 2021 (2nd/3rd prize at DSTC8 Track2 Sub-task1)
  
• W1Label-Efficient Training for Next Response Selection [PDF]
  Seungtaek Choi*, Myeongho Jeong*, Jinyoung Yeo, Seung-won Hwang^† (* equal contribution)
  EMNLP 2020 (workshop, SustaiNLP)
  
• C7Retrieval-Augmented Controllable Review Generation [PDF]
  Jihyeok Kim, Seungtaek Choi, Reinald Kim Amplayo, Seung-won Hwang^†
  COLING 2020
  
• C6Less is More: Attention Supervision with Counterfactuals for Text Classification [PDF]
  Seungtaek Choi, Haeju Park, Jinyoung Yeo, Seung-won Hwang^†
  EMNLP 2020
  
• C5Conditional Response Augmentation for Dialogue using Knowledge Distillation [PDF]
  Myeongho Jeong*, Seungtaek Choi*, Hojae Han, Kyungho Kim, Seung-won Hwang^† (* equal contribution)
  INTERSPEECH 2020
  
• J1Meta-Supervision for Attention using Counterfactual Estimation [PDF]
  Seungtaek Choi, Haeju Park, Seung-won Hwang^†
  DSEJ 2020 (Highly Rated ICDM Issue Invitation)
  
• C4Counterfactual Attention Supervision [PDF]
  Seungtaek Choi, Haeju Park, Seung-won Hwang^†
  ICDM 2019 (short)
  
• C3MICRON: Multigranular Interaction for Contextualizing Representation in Non-factoid Question Answering [PDF]
  Hojae Han*, Seungtaek Choi*, Haeju Park, Seung-won Hwang^† (* equal contribution)
  EMNLP 2019 (short)
  
• C2Visual Choice of Plausible Alternatives: An Evaluation of Image-based Commonsense Causal Reasoning [PDF]
  Jinyoung Yeo, Gyungbok Lee*, Gengyu Wang*, Seungtaek Choi, Hyunsouk Cho, Reinald Kim Amplayo, Seung-won Hwang^† (* equal contribution)
  LREC 2018
  
• C1Machine-translated Knowledge Transfer for Commonsense Causal Reasoning [PDF]
  Jinyoung Yeo, Gengyu Wang, Hyunsouk Cho, Seungtaek Choi, Seung-won Hwang^†
  AAAI 2018
  

Awards & Scholarships

• SustaiNLP registration grant at EMNLP 2020
• Student Travel Grant at INTERSPEECH 2020
• Naver 2019 PhD Fellowship from Naver
• Student Travel Award from ICDM 2019
• Google Travel Grant for ICDM 2019 from Google
• 3rd Place of BIG 2017 CUP Challenge at BIG 2017 conference co-located with WWW 2017
• Top Winner of BIG 2016 CUP Challenge at BIG 2016 conference co-located with WWW 2016
• Computer Science Department Scholarship at Yonsei University 2017-2018

Experience

• Member of Technical Staff, Trillion Labs (Apr 2025 ~ )
• Machine Learning Researcher, Yanolja (Dec 2023 ~ Apr 2025)
• Research Scientist (NLP), Riiid (Mar 2022 ~ Dec 2023)
• Research Internship, Conv AI @ SK T-brain (Aug 2019 ~ Oct 2019)
• Teaching Assitant, Artificial Intelligence @ Yonsei (2017 Spring)