7 papers
Same Concept, Different Directions: Cross-Modal Feature Heterogeneity in Sparse Autoencoders
Chungpa Lee, Jihoon Kwon, Kyle Min +1
Vision-language models map images and text into a joint embedding space. However, these embeddings often entangle multiple semantic features, which limits their interpretability an…
Fine-Tuning Without Forgetting In-Context Learning: A Theoretical Analysis of Linear Attention Models
Chungpa Lee, Jy-yong Sohn, Kangwook Lee
Transformer-based large language models exhibit in-context learning, enabling adaptation to downstream tasks via few-shot prompting with demonstrations. In practice, such models ar…
How to Correctly Report LLM-as-a-Judge Evaluations
Chungpa Lee, Thomas Zeng, Jongwon Jeong +2
Large language models (LLMs) are widely used as scalable evaluators of model responses in lieu of human annotators. However, imperfect sensitivity and specificity of the LLM judges…
On the Similarities of Embeddings in Contrastive Learning
Chungpa Lee, Sehee Lim, Kibok Lee +1
Contrastive learning operates on a simple yet effective principle: Embeddings of positive pairs are pulled together, while those of negative pairs are pushed apart. In this paper,…
A Theoretical Framework for Preventing Class Collapse in Supervised Contrastive Learning
Chungpa Lee, Jeongheon Oh, Kibok Lee +1
Supervised contrastive learning (SupCL) has emerged as a prominent approach in representation learning, leveraging both supervised and self-supervised losses. However, achieving an…
A Generalized Theory of Mixup for Structure-Preserving Synthetic Data
Chungpa Lee, Jongho Im, Joseph H. T. Kim
Mixup is a widely adopted data augmentation technique known for enhancing the generalization of machine learning models by interpolating between data points. Despite its success an…
Analysis of Using Sigmoid Loss for Contrastive Learning
Chungpa Lee, Joonhwan Chang, Jy-yong Sohn
Contrastive learning has emerged as a prominent branch of self-supervised learning for several years. Especially, CLIP, which applies contrastive learning to large sets of captione…