Person
Intel Labs
2 stories · sorted newest first · 📡 RSS
Intel Labs and Hugging Face Develop Dynamic Speculative Decoding for Faster Text Generation
A new method, called dynamic speculative decoding, can speed up text generation by up to 2.7x. It's now the default operational mo
Intel Labs and Hugging Face Introduce Universal Assisted Generation for Efficient Text Generation
The new UAG method enables assisted generation to work with any pair of target and assistant models, accelerating inference by up