Zobrazeno 1 - 10
of 218
pro vyhledávání: '"Metze, Florian"'
Quantization has become a predominant approach for model compression, enabling deployment of large models trained on GPUs onto smaller form-factor devices for inference. Quantization-aware training (QAT) optimizes model parameters with respect to the
Externí odkaz:
http://arxiv.org/abs/2212.05603
Autor:
Park, Yookoon, Azab, Mahmoud, Xiong, Bo, Moon, Seungwhan, Metze, Florian, Kundu, Gourab, Ahmed, Kirmani
Cross-modal contrastive learning has led the recent advances in multimodal retrieval with its simplicity and effectiveness. In this work, however, we reveal that cross-modal contrastive learning suffers from incorrect normalization of the sum retriev
Externí odkaz:
http://arxiv.org/abs/2212.11790
Autor:
Arora, Siddhant, Dalmia, Siddharth, Yan, Brian, Metze, Florian, Black, Alan W, Watanabe, Shinji
End-to-end spoken language understanding (SLU) systems are gaining popularity over cascaded approaches due to their simplicity and ability to avoid error propagation. However, these systems model sequence labeling as a sequence prediction task causin
Externí odkaz:
http://arxiv.org/abs/2210.15734
Quantization is an effective technique to reduce memory footprint, inference latency, and power consumption of deep learning models. However, existing quantization methods suffer from accuracy degradation compared to full-precision (FP) models due to
Externí odkaz:
http://arxiv.org/abs/2210.07171
Autor:
Yan, Brian, Dalmia, Siddharth, Higuchi, Yosuke, Neubig, Graham, Metze, Florian, Black, Alan W, Watanabe, Shinji
Connectionist Temporal Classification (CTC) is a widely used approach for automatic speech recognition (ASR) that performs conditionally independent monotonic alignment. However for translation, CTC exhibits clear limitations due to the contextual an
Externí odkaz:
http://arxiv.org/abs/2210.05200
Most recent speech recognition models rely on large supervised datasets, which are unavailable for many low-resource languages. In this work, we present a speech recognition pipeline that does not require any audio for the target language. The only a
Externí odkaz:
http://arxiv.org/abs/2209.02842
Autor:
Huang, Po-Yao, Xu, Hu, Li, Juncheng, Baevski, Alexei, Auli, Michael, Galuba, Wojciech, Metze, Florian, Feichtenhofer, Christoph
This paper studies a simple extension of image-based Masked Autoencoders (MAE) to self-supervised representation learning from audio spectrograms. Following the Transformer encoder-decoder design in MAE, our Audio-MAE first encodes audio spectrogram
Externí odkaz:
http://arxiv.org/abs/2207.06405
Autor:
Dalmia, Siddharth, Okhonko, Dmytro, Lewis, Mike, Edunov, Sergey, Watanabe, Shinji, Metze, Florian, Zettlemoyer, Luke, Mohamed, Abdelrahman
State-of-the-art encoder-decoder models (e.g. for machine translation (MT) or automatic speech recognition (ASR)) are constructed and trained end-to-end as an atomic unit. No component of the model can be (re-)used without the others, making it impos
Externí odkaz:
http://arxiv.org/abs/2206.03318
Autor:
Palaskar, Shruti, Bhagia, Akshita, Bisk, Yonatan, Metze, Florian, Black, Alan W, Marasović, Ana
Combining the visual modality with pretrained language models has been surprisingly effective for simple descriptive tasks such as image captioning. More general text generation however remains elusive. We take a step back and ask: How do these model
Externí odkaz:
http://arxiv.org/abs/2205.11686
Traditionally, in Audio Recognition pipeline, noise is suppressed by the "frontend", relying on preprocessing techniques such as speech enhancement. However, it is not guaranteed that noise will not cascade into downstream pipelines. To understand th
Externí odkaz:
http://arxiv.org/abs/2205.03268