Zobrazeno 1 - 10
of 495
pro vyhledávání: '"Kumar, Shankar"'
Autor:
Velikovich, Leonid, Li, Christopher, Caseiro, Diamantino, Kumar, Shankar, Rondon, Pat, Joshi, Kandarp, Velez, Xavier
For end-to-end Automatic Speech Recognition (ASR) models, recognizing personal or rare phrases can be hard. A promising way to improve accuracy is through spelling correction (or rewriting) of the ASR lattice, where potentially misrecognized phrases
Externí odkaz:
http://arxiv.org/abs/2409.16469
One challenge in speech translation is that plenty of spoken content is long-form, but short units are necessary for obtaining high-quality translations. To address this mismatch, we adapt large language models (LLMs) to split long ASR transcripts in
Externí odkaz:
http://arxiv.org/abs/2310.13678
Federated Averaging, and many federated learning algorithm variants which build upon it, have a limitation: all clients must share the same model architecture. This results in unused modeling capacity on many clients, which limits model performance.
Externí odkaz:
http://arxiv.org/abs/2310.02549
Autor:
Zhu, Yun, Liu, Yinxiao, Stahlberg, Felix, Kumar, Shankar, Chen, Yu-hui, Luo, Liangchen, Shu, Lei, Liu, Renjie, Chen, Jindong, Meng, Lei
Large Language Models (LLMs) have demonstrated impressive capabilities for text rewriting. Nonetheless, the large sizes of these models make them impractical for on-device inference, which would otherwise allow for enhanced privacy and economical inf
Externí odkaz:
http://arxiv.org/abs/2308.11807
We propose a method of segmenting long-form speech by separating semantically complete sentences within the utterance. This prevents the ASR decoder from needlessly processing faraway context while also preventing it from missing relevant context wit
Externí odkaz:
http://arxiv.org/abs/2305.18419
Autor:
Carey, CJ, Dick, Travis, Epasto, Alessandro, Javanmard, Adel, Karlin, Josh, Kumar, Shankar, Medina, Andres Munoz, Mirrokni, Vahab, Nunes, Gabriel Henrique, Vassilvitskii, Sergei, Zhong, Peilin
Compact user representations (such as embeddings) form the backbone of personalization services. In this work, we present a new theoretical framework to measure re-identification risk in such user representations. Our framework, based on hypothesis t
Externí odkaz:
http://arxiv.org/abs/2304.07210
A challenge in spoken language translation is that plenty of spoken content is long-form, but short units are necessary for obtaining high-quality translations. To address this mismatch, we fine-tune a general-purpose, large language model to split l
Externí odkaz:
http://arxiv.org/abs/2212.09895
We report on novel investigations into training models that make sentences concise. We define the task and show that it is different from related tasks such as summarization and simplification. For evaluation, we release two test sets, consisting of
Externí odkaz:
http://arxiv.org/abs/2211.04126
Recent trends towards training ever-larger language models have substantially improved machine learning performance across linguistic tasks. However, the huge cost of training larger models can make tuning them prohibitively expensive, motivating the
Externí odkaz:
http://arxiv.org/abs/2209.04683
Autor:
Malmi, Eric, Dong, Yue, Mallinson, Jonathan, Chuklin, Aleksandr, Adamek, Jakub, Mirylenka, Daniil, Stahlberg, Felix, Krause, Sebastian, Kumar, Shankar, Severyn, Aliaksei
Text-editing models have recently become a prominent alternative to seq2seq models for monolingual text-generation tasks such as grammatical error correction, simplification, and style transfer. These tasks share a common trait - they exhibit a large
Externí odkaz:
http://arxiv.org/abs/2206.07043