Kichang Yang

PhD student, Seoul National University

1 paper at NeurIPS 2025

OpenReview· Semantic Scholar· Google Scholar

Poster Session 6

1 paper

Friday, December 5, 2025 · 4:30 PM → 7:30 PM

Exhibit Hall C,D,E

VLM in a flash: I/O-Efficient Sparsification of Vision-Language Model via Neuron Chunking

#4519 · Kichang Yang, Seonjun Kim, Minjae Kim, Nairan Zhang, chi zhang, Youngki Lee

Neuron Chunking is a hardware-aware sparsification framework that abstracts access patterns into contiguity distributions to couple neuron selection with flash I/O behavior and improve I/O efficiency in VLM inference.