一位用户在推文上发布了一项新的子二次注意力技术,声称可使长上下文大语言模型成本降低10倍且不牺牲性能,并附有链接。该技术可能影响AI模型的效率。
"Introducing a breakthrough new technique for sub-quadratic attention, making long-context LLMs 10x cheaper without sacrificing performance"
Me: https://t.co/IkkOwgBy3k
likes: 57 | retweets: 3 | replies: 2 | views: 4158