We develop a procedure for Int8 matrix multiplication for feed-forward and attention projection layers in transformers, which cut the memory needed for inference by half while retaining full precision performance.
Ranked #2 on Language Modelling on C4
In order to reap the benefits and avoid the drawbacks of CBFT and CFFT, we propose a novel framework with a Hybrid Feature Transformation module (HFT).
Context clusters (CoCs) view an image as a set of unorganized points and extract features via simplified clustering algorithm.
We present OpenSeeD, a simple Open-vocabulary Segmentation and Detection framework that jointly learns from different segmentation and detection datasets.
Ranked #1 on Instance Segmentation on ADE20K val (using extra training data)
We study the capabilities of speech processing systems trained simply to predict large amounts of transcripts of audio on the internet.
Ranked #1 on Speech Recognition on CHiME6
We show that intermediate self-attention maps of a masked generative transformer encode important structural information of the input image, such as scene layout and object shape, and we propose a novel sampling method based on this observation to enable structure-guided generation.
Deep supervision, which involves extra supervisions to the intermediate features of a neural network, was widely used in image classification in the early deep learning era since it significantly reduces the training difficulty and eases the optimization like avoiding gradient vanish over the vanilla training.
Scenic: A Jax Library for Computer Vision Research and Beyond
A critical enabler for progress in neuromorphic computing research is the ability to transparently evaluate different neuromorphic solutions on important tasks and to compare them to state-of-the-art conventional solutions.
In this report, we present a fast and accurate object detection method dubbed DAMO-YOLO, which achieves higher performance than the state-of-the-art YOLO series.
Ranked #15 on Real-Time Object Detection on COCO