Why Are Linear RNNs More Parallelizable?
arXiv:2603.03612v1 Announce Type: new Abstract: The community is increasingly exploring linear RNNs (LRNNs) as language models, motivated by their expressive power and parallelizability. While prior …
William Merrill, Hongjian Jiang, Yanhong Li, Ashish Sabharwal
조회수 43회