5 papers
Compliance, Capability, and Conflict: Benchmarking Multimodal LLMs under System Messages
Juan Yeo, Geewook Kim
Production deployments of Multimodal Large Language Models (MLLMs) increasingly rely on system messages to govern model behavior. Yet existing benchmarks either evaluate constraint…
Stable On-Policy Distillation through Adaptive Target Reformulation
Ijun Jang, Jewon Yeom, Juan Yeo +2
Knowledge distillation (KD) is a widely adopted technique for transferring knowledge from large language models to smaller student models; however, conventional supervised KD often…
Patch-Level Kernel Alignment for Dense Self-Supervised Learning
Juan Yeo, Ijun Jang, Taesup Kim
Dense self-supervised learning (SSL) methods showed its effectiveness in enhancing the fine-grained semantic understandings of vision models. However, existing approaches often rel…
ATAS: Any-to-Any Self-Distillation for Enhanced Open-Vocabulary Dense Prediction
Juan Yeo, Soonwoo Cha, Jiwoo Song +2
Vision-language models such as CLIP have recently propelled open-vocabulary dense prediction tasks by enabling recognition of a broad range of visual concepts. However, CLIP still…
When Vision Models Meet Parameter Efficient Look-Aside Adapters Without Large-Scale Audio Pretraining
Juan Yeo, Jinkwan Jang, Kyubyung Chae +2
Recent studies show that pretrained vision models can boost performance in audio downstream tasks. To enhance the performance further, an additional pretraining stage with large sc…