Quantization-aware and Tensor-compressed Training of Transformers for Natural Language Understanding.
Zi YangSamridhi ChoudharySiegfried KunzmannZheng ZhangPublished in: INTERSPEECH (2023)
Keyphrases
- natural language understanding
- text understanding
- semantic analysis
- natural language
- knowledge representation
- natural language processing
- semantic representations
- dialogue system
- language understanding
- spoken dialog systems
- training set
- joint inference
- high order
- huffman coding
- data structure
- quantization error
- diffusion tensor
- information retrieval