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
GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher
Graph of Thoughts: Solving Elaborate Problems with Large Language Models
Platypus: Quick, Cheap, and Powerful Refinement of LLMs
FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterization
Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP
Automatically Correcting Large Language Models: Surveying the landscape of diverse self-correction strategies
Simple synthetic data reduces sycophancy in large language models
Shepherd: A Critic for Language Model Generation
AgentBench: Evaluating LLMs as Agents
MetaGPT: Meta Programming for Multi-Agent Collaborative Framework
Revisiting the Minimalist Approach to Offline Reinforcement Learning
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
LISA: Reasoning Segmentation via Large Language Model
Scaling Relationship on Learning Mathematical Reasoning with Large Language Models
The All-Seeing Project: Towards Panoptic Visual Recognition and Understanding of the Open World
ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs
Universal and Transferable Adversarial Attacks on Aligned Language Models
FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarios
Aligning Large Language Models with Human: A Survey
FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Sets
Join Podbean Ads Marketplace and connect with engaged listeners.
Advertise Today
Create your
podcast in
minutes
It is Free
WSJ Tech News Briefing
The 404 Media Podcast
Rebel Tech
CyberWire Daily
Click Here