Recent advancements in audio generation have been spurred by the evoluti...
In language modeling based music generation, a generated waveform is
rep...
Automatic Speech Recognition (ASR) models need to be optimized for speci...
Knowing exactly how many data points need to be labeled to achieve a cer...
Weight-sharing supernet has become a vital component for performance
est...
Several post-training quantization methods have been applied to large
la...
Fusing camera with LiDAR is a promising technique to improve the accurac...
Emerging real-time multi-model ML (RTMM) workloads such as AR/VR and dro...
Diffusion models have emerged as a powerful tool for point cloud generat...
Real-time multi-model multi-task (MMMT) workloads, a new form of deep
le...
This paper proposes a hardware-efficient architecture, Linearized Convol...
Efficient deep neural network (DNN) models equipped with compact operato...
Streaming ASR with strict latency constraints is required in many speech...
Low-rank tensor compression has been proposed as a promising approach to...
Vision transformers (ViTs) have attracted much attention for their super...
From wearables to powerful smart devices, modern automatic speech recogn...
Automatic speech recognition (ASR) has become increasingly ubiquitous on...
On-device speech recognition requires training models of different sizes...
Introducing the transformer structure into computer vision tasks holds t...
Deep learning-based RAW image denoising is a quintessential problem in i...
Modern day computing increasingly relies on specialization to satiate gr...
Recurrent transducer models have emerged as a promising solution for spe...
Weight-sharing neural architecture search (NAS) is an effective techniqu...
Low-precision deep neural network (DNN) training has gained tremendous
a...
Video transmission applications (e.g., conferencing) are gaining momentu...
Scale variance among different sizes of body parts and objects is a
chal...
Leveraging temporal information has been regarded as essential for devel...
Data augmentation (DA) is an essential technique for training
state-of-t...
Neural architecture search (NAS) has shown great promise designing
state...
Powerful yet complex deep neural networks (DNNs) have fueled a booming d...
Weight quantization for deep ConvNets has shown promising results for
ap...
Neural Architecture Search (NAS) automates and prospers the design of ne...
Neural network accelerator is a key enabler for the on-device AI inferen...
Neural Architecture Search (NAS) has demonstrated its power on various A...
Personalized recommendation systems leverage deep learning models and ac...
Designing energy-efficient networks is of critical importance for enabli...
Recent advances in deep neural networks (DNNs) have made DNNs the backbo...
Federated learning enables resource-constrained edge compute devices, su...
Deep Neural Networks are becoming increasingly popular in always-on IoT ...
Efficient and compact neural network models are essential for enabling t...
Hardware acceleration of Deep Neural Networks (DNNs) aims to tame their
...
Keyword spotting (KWS) is a critical component for enabling speech based...
Deep convolutional neural network (CNN) inference requires significant a...