Publications

Conference Publications

Breaking the Lock-in: Diversifying Text-to-Image Generation via Representation Modulation

[paper]

Dahee Kwon, Haeun Lee and Jaesik Choi

ICML 2026

SPICE: Simple Polysemantic Feature Interpretation via Clustering-based Explanations

Sehyun Lee, Dahee Kwon, Damin Lee and Jaesik Choi

ECCV 2026

Granular Concept Circuits: Toward a Fine-Grained Circuit Discovery for Concept Representations

[paper]

Dahee Kwon*, Sehyun Lee* and Jaesik Choi

* equally contributed

ICCV 2025

Enhancing Creative Generation on Stable Diffusion-based Models

[paper]

Jiyeon Han*, Dahee Kwon* and Jaesik Choi

* equally contributed

CVPR 2025

Understanding Distributed Representations of Concepts in Deep Neural Networks without Supervision

[paper]

Wonjoon Chang*, Dahee Kwon* and Jaesik Choi

* equally contributed

AAAI 2024 (Oral)

Under Review

Causal Path Tracing in Transformers

Won Jo, Dahee Kwon, Jongeun Baek, Cheongwoong Kang and Jaesik Choi

Under Review, ICML 2026

Preprints

MAC: Memory-Augmented Contrastive Learning for Time Series Anomaly Detection

Dahee Kwon, Enver Menadjiev, Qin Xie and Jaesik Choi

Preprint

Dual Masking for Domain Generalization

Dahee Kwon and Jaesik Choi

Preprint