Latest Research Highlight

Dementia-R1: Reinforced Pretraining and Reasoning from Unstructured Clinical Notes for Real-World Dementia Prognosis

Choonghan Kim et al., EMNLP 2026

Dementia-R1: Reinforced Pretraining and Reasoning from Unstructured Clinical Notes for Real-World Dementia Prognosis

News

Aug 22, 2026 Congratulations to BISPL's New Tenure-Track Faculty
Aug 21, 2026 Three papers (including two in Findings) accepted to EMNLP 2026!
Jul 13, 2026 BISPL Students Awarded Funded Research Internships at Prestigious Research Institutes
Jun 18, 2026 Congratulations! Five papers accepted to ECCV 2026.
May 13, 2026 This Semester’s Global Intern Dispatch from BISPL

About BISPL

BISPL —the Bio Imaging, Signal Processing, and Machine Learning Lab at KAIST AI — is led by Prof. Jong Chul Ye, an IEEE Fellow. Our research spans theory & reinforcement Learning (RL) for diffusion/flow Models, LLMs & agentic AI, vision-language-action (VLA) Models, world models, computer vision, AI for science & healthcare, and medical imaging.

Across these areas, we seek to uncover the mathematical beauty and fundamental principles of modern AI and translate them into real-world impact. We ask not only how AI works, but why it works, developing new theories and algorithms from challenging practical problems. Building on deep expertise in biomedical imaging, signal processing, inverse problems, and machine learning, we turn real-world applications into sources of fundamental insight—and fundamental advances into transformative technologies.

Explore our people, publications, and latest news.

Most Recent Publications

  1. [C103] Dementia-R1: Reinforced Pretraining and Reasoning from Unstructured Clinical Notes for Real-World Dementia Prognosis
    Choonghan Kim, Hyunmin Hwang, Hangeol Chang, Jaemin Kim, Jinse Park, Jae-Sung Lim, and Jong Chul Ye
    In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, Oct 2026
  2. [C102] Adaptive Guidance for Retrieval-Augmented Masked Diffusion Models
    Jaemin Kim and Jong Chul Ye
    In Findings of the Association for Computational Linguistics: EMNLP 2026, Oct 2026
  3. [C101] PACE-RAG: Patient-Aware Contextual and Evidence-Constrained RAG for Clinical Drug Recommendation
    Chaeyoung Huh, Hyunmin Hwang, Jung Hwan Shin, Sungyang Jo, Jinse Park, and Jong Chul Ye
    In Findings of the Association for Computational Linguistics: EMNLP 2026, Oct 2026
  4. [C100] FILT3R: Latent State Adaptive Kalman Filter for Streaming 3D Reconstruction
    Seonghyun Jin and Jong Chul Ye
    In The 19th European Conference on Computer Vision (ECCV), 2026
    Bib
  5. [C99] InverseCrafter: Efficient Video ReCapture as a Latent Domain Inverse Problem
    Yeobin Hong, Suhyeon Lee, Hyungjin Chung, and Jong Chul Ye
    In The 19th European Conference on Computer Vision (ECCV), 2026
    Bib
  6. [C98] Tiled Prompts: Overcoming Prompt Misguidance in Image and Video Super-Resolution
    Bryan Sangwoo Kim, Jonghyun Park, and Jong Chul Ye
    In The 19th European Conference on Computer Vision (ECCV), 2026
    Bib