Back to Search
Start Over
DOTA: detect and omit weak attentions for scalable transformer acceleration
- Source :
- Proceedings of the 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems.
- Publication Year :
- 2022
- Publisher :
- ACM, 2022.
Details
- Database :
- OpenAIRE
- Journal :
- Proceedings of the 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems
- Accession number :
- edsair.doi...........3d39063d628436e1716e84fbc5b6fe7c