avatar

Wonduk Seo

Seoul, South Korea
Bachelor in Information Management, Peking University
Enhans AI
1duck2dpdy@gmail.com


About Me

Hi! I’m Wonduk Seo, an AI Research Scientist at Enhans AI since January 2025 and a 2025 graduate of Peking University (Bachelor in Information Management, Big Data Management & Application). Previously, I was advised by Prof. Yi Bu at Peking University. I’m passionate about building scalable, transparent, and interpretable methods that keep LLMs grounded in human values, sustain behavioral consistency, and mitigate hallucinations in real-world deployments.

Research Interests

News

  • [Dec. 2025] One first author paper got accepted to ACM International Conference on Web Search and Data Mining (WSDM)!
  • [Nov. 2025] One co-first author paper got accepted to ACM/IEEE Joint Conference on Digital Libraries (JCDL).
  • [Oct. 2025] One co-first author paper got accepted to IEEE International Conference on Big Data 2025 Industry Track for Oral Presentation 🇲🇴.
  • [Jul. 2025] One first author paper got accepted to AAAI/ACM Conference on AI, Ethics, and Society (AIES) Main Track 🇪🇸.
  • [Jul. 2025] Graduated from Peking University (Bachelor in Information Management, Big Data Management & Application) 🎉!
  • [Jun. 2025] One first author paper accepted for Oral Presentation at the ACM SIGKDD 2025 Workshop on AI Agent for Information Retrieval 🇨🇦.
  • [Apr. 2025] One co-first author paper got accepted to IEEE Engineering in Medicine and Biology Society (EMBC) for Oral Presentation 🇩🇰.
  • [Apr. 2025] Gave a poster talk at the International Conference on Research in Computational Molecular Biology (RECOMB) 🇰🇷.
  • [Feb. 2025] One first author paper accepted by Quantitative Science Studies (QSS).
  • [Jan. 2025] Concluded my role as an LLM researcher at Baidu Inc..
  • [Jan. 2025] Rejoined Enhans AI as an AI Research Scientist.
  • [Nov. 2024] Ranked 9th in the ACM'24 FinanceRAG Challenge as Team Leader.
  • [May. 2024] 🥉 Earned a bronze medal in Kaggle’s Enefit Predict Energy Behavior of Prosumers (Solo).
  • [Apr. 2024] 🥉 Earned a bronze medal in Kaggle’s Harvard Medical School Brain Activity Classification (Team Leader).
  • [Feb. 2024] Wrapped up my role as a Machine Learning Engineer at Enhans AI to pursue research-focused opportunities.
  • [Dec. 2023] 🥈 Earned a silver medal in Kaggle’s CAFA 5 Protein Function Prediction (Solo).
  • [Oct. 2023] 🥉 Earned a bronze medal in Kaggle’s LLM Science Exam (Team).
  • [Aug. 2023] Completed my Data Scientist internship at LG Chem.
  • [Apr. 2023] Completed a Data Scientist internship at BetaBrain.
  • [Apr. 2022] Began a long-term collaboration with Prof. Yi Bu.
  • [Dec. 2020] Began my AI journey, diving into machine learning research and engineering.

Publications

Accepted Papers

  1. WSDM
    A New Query Expansion Approach for Enhancing Information Retrieval via Agent-Mediated Dialogic Inquiry
    Wonduk Seo, Hyunjin An, Seunghyun Lee
    ACM International Conference on Web Search and Data Mining (WSDM), 2026.
    ACM SIGKDD Workshop on AI Agent for Information Retrieval (SIGKDD-Agent4IR), 2025. Oral Presentation

  2. AAAI-AIES
    ValuesRAG: Enhancing Cultural Alignment Through Retrieval-Augmented Contextual Learning
    Wonduk Seo, Zonghao Yuan, Yi Bu
    Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (AAAI-AIES), 2025.

  3. BigData
    Question-to-Knowledge: Multi-Agent Generation of Inspectable Facts for Product Mapping
    Wonduk Seo*, Taesub Shin*, Hyunjin An, Dokyun Kim, Seunghyun Lee
    2025 IEEE International Conference on Big Data (BigData) Industry Track, 2025. Oral Presentation

  4. JCDL
    Better by Comparison: Retrieval-Augmented Contrastive Reasoning for Automatic Prompt Optimization
    Juhyeon Lee*, Wonduk Seo*, Hyunjin An, Seunghyun Lee, Yi Bu
    Proceedings of the 25th ACM/IEEE Joint Conference on Digital Libraries (JCDL), Short Paper, 2025.

  5. EMBC
    Pesti-Gen: Unleashing a Generative Molecule Approach for Toxicity Aware Pesticide Design
    Taehan Kim*, Wonduk Seo*
    47th International Conference of the IEEE Engineering in Medicine and Biology (EMBC), 2025. Oral Presentation

  6. QSS
    Transforming Role Classification in Scientific Teams using LLMs and Advanced Predictive Analytics
    Wonduk Seo, Yi Bu
    Quantitative Science Studies (QSS), 2025.

Working Papers

  1. W1
    Prompt Optimization via Retrieved Reasoning Assets and Multi-Agent Analysis
    Wonduk Seo, Juhyeon Lee, Junseo Koh, Hyunjin An, Jian Park, Seunghyun Lee, Haihua Chen, Yi Bu
    Under review.

  2. W2
    Automated Visualization Code Synthesis via Multi-Path Reasoning and Feedback-Driven Optimization
    Wonduk Seo, Daye Kang, Hyunjin An, Seunghyun Lee, Taehan Kim, Zonghao Yuan, Yi Bu
    Under review.

  3. W3
    MARIC: Multi-Agent Reasoning for Image Classification
    Wonduk Seo*, Minhyeong Yu*, Hyunjin An, Seunghyun Lee
    Under review.

  4. W4
    SPIO: Ensemble and Selective Strategies via LLM-Based Multi-Agent Planning in Automated Data Science
    Wonduk Seo*, Juhyeon Lee*, Yi Bu
    Under revision.

  5. W5
    GenCRF: Generative Clustering and Reformulation Framework for Enhanced Intent-Driven Information Retrieval
    Wonduk Seo et al.
    Under revision.

  6. W6
    AIAP: A Multi Agent Workflow Builder with Visual Programming and Natural Language Prompts
    Hyunjin An, Yongwon Kim, Wonduk Seo, Jian Park, Daye Kang, Changhoon Oh, Dokyun Kim, Seunghyun Lee
    Under review.

Competition Awards

Ranked 9th, ACM'24 FinanceRAG Challenge (Team Leader), 11/2024

🥉 Kaggle – Enefit Predict Energy Behavior of Prosumers (Solo), 05/2024

🥉 Kaggle – Harvard Medical School Brain Activity Classification (Team Leader), 04/2024

🥈 Kaggle – CAFA 5 Protein Function Prediction (Solo), 12/2023

🥉 Kaggle – LLM Science Exam (Team), 10/2023

Dacon – Jeju Product Price Prediction (Solo; Top 1% among 1.4K participants), 10/2023

Advisors and Collaborators

Advisors

Collaborators

Mentees


Powered by Jekyll and Minimal Light theme.