2 papers
cs.CV2026
Keep the Needle, Prune the Haystack: Defect-Preserving Token Pruning for Efficient Zero-Shot Anomaly Detection
Yanning Hou, Jingyuan Zhang, Xiaoyun Wang +3
Zero-shot visual anomaly detection has achieved remarkable progress, with recent vision-only approaches further improving performance while simplifying the inference pipeline. Howe…
cs.LG2026
EPTS: Elastic Post-Training Sparsity for Efficient Large Language Model Compression
Ke Xu, Jiaqi Wan, Wenhao Hu +2
Post-Training Sparsity (PTS) has emerged as a crucial paradigm for compressing Large Language Models to facilitate efficient deployment on resource-constrained devices. However, ex…