一作ICML 2026Poster· Deep Learning / Large Language Models
ReSpinQuant: Efficient Layer-Wise LLM Quantization via Subspace Residual Rotation Approximation
Suyoung Kim · Sunghyun Wee · Hyeonjin Kim · Kyomin Hwang · Hyunho Lee · NOJUN KWAK
效率与系统OpenReview