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
LCM-LoRA: A Universal Stable-Diffusion Acceleration Module
CogVLM: Visual Expert for Pretrained Language Models
How Can Recommender Systems Benefit from Large Language Models: A Survey
Lumos: Learning Agents with Unified Data, Modular Design, and Open-Source LLMs
Generative Pretraining in Multimodality
Evaluating Large Language Models: A Comprehensive Survey
LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression
iTransformer: Inverted Transformers Are Effective for Time Series Forecasting
Zephyr: Direct Distillation of LM Alignment
Large Language Models for Software Engineering: Survey and Open Problems
Eureka: Human-Level Reward Design via Coding Large Language Models
Zero123++: a Single Image to Consistent Multi-view Diffusion Base Model
Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference
AgentTuning: Enabling Generalized Agent Abilities for LLMs
MemGPT: Towards LLMs as Operating Systems
AgentTuning: Enabling Generalized Agent Abilities for LLMs
Character-LLM: A Trainable Agent for Role-Playing
OpenAgents: An Open Platform for Language Agents in the Wild
Text Embeddings Reveal (Almost) As Much As Text
Ferret: Refer and Ground Anything Anywhere at Any Granularity
Join Podbean Ads Marketplace and connect with engaged listeners.
Advertise Today
Create your
podcast in
minutes
It is Free
WSJ Tech News Briefing
CyberWire Daily
The WAN Show
gm! crypto
Techmeme Ride Home