DQMix-BERT: Distillation-aware Quantization with Mixed Precision for BERT Compression.
Yan TanLei JiangPeng ChenChaodong TongPublished in: SMC (2023)
Keyphrases
- quantization noise
- data compression
- compression scheme
- lossy image compression
- compression ratio
- efficient compression
- uniform quantization
- huffman coding
- high precision
- image compression
- compression algorithm
- wavelet image coding
- lookup table
- entropy coding
- lossy compression
- average precision
- transform coding
- scalar quantization
- compressed data
- adaptive quantization
- compression rate
- precision and recall