ReAT: Long Text Encoder Representations with Attention Tokens.
Harunori KawanoSota ShimizuPublished in: SII (2023)
Keyphrases
- semantic representations
- text documents
- rate distortion
- automatically discovering
- free text
- database
- bit rate
- text retrieval
- symbolic representation
- low complexity
- motion estimation
- line segments
- power consumption
- image quality
- natural language processing
- keywords
- text data
- textual data
- video coding
- natural language generation
- neural network
- text classification