paper-conference

Revisiting Layer-level Residual Connections for Efficient Object Detection
This paper proposed a rewiring scheme for effectively compressing YOLOv8.
Bespoke: A Block-Level Neural Network Optimization Framework for Low-Cost Deployment
This paper proposed a framework for compressing neural network models with consideration of their target platform.
PDF
Rethinking Group Fisher Pruning for Efficient Label-Free Network Compression
This paper proposed an efficient channel pruning without labels based on Group Fisher Pruning.
PDF
Block-wise Word Embedding Compression Revisited: Better Weighting and Structuring
This paper proposed a compression method for word embedding based on block-wise low-rank compression.
PDF DOI
A Fast Approximation for Influence Maximization in Large Social Networks
This paper proposed an efficient approximation algorithm for influence maximization. This algorithms utilizes the two-hop neighbors of each node to evaluate expected influence spread.
PDF