4 papers
Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts
Seungmin Oh, Seunghun Kang, Jongbin Ryu
Vision-language models (VLMs) achieve strong zero-shot transferability but remain vulnerable to target-domain shifts at inference time. Test-time adaptation (TTA) offers a practica…
Layer-wise Curriculum Learning for Efficient LLM Compression
Donggeon Lee, Dooyeon Na, Seungmin Oh +1
In this paper, we introduce layer-wise curriculum learning for efficient LLM compression. The proposed method facilitates the knowledge transfer from the teacher model to the stude…
Train Overcomplete, Deploy Compact: Scaling Recovery Capacity for Structured LLM Pruning
Seungmin Oh, Donggeon Lee, Jongbin Ryu
Large language models achieve strong performance across diverse tasks, but deployment remains costly because of memory, latency, and energy demands. Structured pruning reduces thes…
Re-calibrated Contrastive Loss for Transformation-Aware Prompt Conditioning in Vision-Language Models
Seungmin Oh, Seunghun Kang, Jongbin Ryu
Ensuring effective transfer learning for vision-language models without compromising their generalization performance is crucial. However, many existing methods overlook data chara…