Haas-ja-v1.0

Papers using or citing this model 3

• Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition 224 citations · 2023 · arXiv (Cornell University) · arXiv

  Conformer-based models have become the dominant end-to-end architecture for speech processing tasks. With the objective of enhancing the conformer architecture for efficient training and inference, we carefully redesigned Conformer with...

• Kimi Linear: An Expressive, Efficient Attention Architecture 204 citations · 2025 · arXiv (Cornell University) · arXiv

  We introduce Kimi Linear, a hybrid linear attention architecture that, for the first time, outperforms full attention under fair comparisons across various scenarios -- including short-context, long-context, and reinforcement learning...

• Efficient Sequence Transduction by Jointly Predicting Tokens and Durations 91 citations · 2023 · arXiv (Cornell University) · arXiv

  This paper introduces a novel Token-and-Duration Transducer (TDT) architecture for sequence-to-sequence tasks. TDT extends conventional RNN-Transducer architectures by jointly predicting both a token and its duration, i.e. the number of...