Deploying convolutional neural networks (CNNs) on embedded devices is difficult due to the limited memory and computation resources. The redundancy in feature maps is an important characteristic of those successful CNNs, but has rarely been investigated in neural architecture design. This paper proposes a novel Ghost module to generate more feature maps from cheap operations. Based on a set of intrinsic feature maps, we apply a series of linear transformations with cheap cost to generate many ghost feature maps that could fully reveal information underlying intrinsic features.
2019: Kai Han, Yunhe Wang, Qi Tian, Jianyuan Guo, Chunjing Xu, Chang Xu
https://arxiv.org/pdf/1911.11907v2.pdf
LoRA: Low-Rank Adaptation of Large Language Models
Pretraining Without Attention
ImageBind: One Embedding Space To Bind Them All
ZipIt! Merging Models from Different Tasks without Training
Chain of Thought Prompting Elicits Reasoning in Large Language Models
CodeGen2: Lessons for Training LLMs on Programming and Natural Languages
Shap-E: Generating Conditional 3D Implicit Functions
OPT: Open Pre-trained Transformer Language Models
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions
Large Language Models Can Self-Improve
Text-to-Audio Generation using Instruction-Tuned LLM and Latent Diffusion Model
WizardLM: Empowering Large Language Models to Follow Complex Instructions
Harnessing the Power of LLMs in Practice: A Survey on ChatGPT and Beyond
Track Anything: Segment Anything Meets Videos
AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Head
DINOv2: Learning Robust Visual Features without Supervision
Robust Speech Recognition via Large-Scale Weak Supervision
High-Resolution Image Synthesis with Latent Diffusion Models
Segment Everything Everywhere All at Once
SadTalker: Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animation
Join Podbean Ads Marketplace and connect with engaged listeners.
Advertise Today
Create your
podcast in
minutes
It is Free
Cyber Security Headlines
The WAN Show
Software Engineering Daily
Babbage from The Economist
Cybersecurity Today