Showing cs.LGShow all
2 papers · 1 filter
cs.LG2025
Beyond One-Way Pruning: Bidirectional Pruning-Regrowth for Extreme Accuracy-Sparsity Tradeoff
Junchen Liu, Yi Sheng
As a widely adopted model compression technique, model pruning has demonstrated strong effectiveness across various architectures. However, we observe that when sparsity exceeds a…
cs.LG2025
SmallThinker: A Family of Efficient Large Language Models Natively Trained for Local Deployment
Yixin Song, Zhenliang Xue, Dongliang Wei +11
While frontier large language models (LLMs) continue to push capability boundaries, their deployment remains confined to GPU-powered cloud infrastructure. We challenge this paradig…