Qrange-group

All

6 repositories

Mirror-Gradient
Public
WWW'24, Mirror Gradient (MG) makes multimodal recommendation models approach flat local minima easier compared to models with normal training.
recommender-system recommendation multimodal gradient-method flat-local-minima
Python
•
MIT License
•1•11•0•0•Updated Nov 1, 2024Nov 1, 2024
SUR-adapter
Public
ACM MM'23 (oral), SUR-adapter for pre-trained diffusion models can acquire the powerful semantic understanding and reasoning capabilities from large language models to build a high-quality textual semantic representation for text-to-image generation.
adapter pytorch image-generation knowledge-distillation diffusion-models large-language-models
Python
•
MIT License
•2•117•8•0•Updated Apr 24, 2024Apr 24, 2024
SEM
Public
SEM can automatically decide to select and integrate attention operators to compute attention maps.
pytorch image-classification self-attention excitation-networks
Python
•
MIT License
•2•9•0•0•Updated Jun 16, 2023Jun 16, 2023
SPEM
Public
MMM'23, SPEM adopts a self-adaptive pooling strategy based on global max-pooling, global min-pooling and a lightweight module for producing the attention map.
pytorch image-classification pooling self-attention
Python
•
MIT License
•0•2•0•0•Updated Jun 16, 2023Jun 16, 2023
LSAS
Public
ICME'23, Lightweight sub-attention strategy (LSAS) utilizes high-order sub-attention modules to improve the original self-attention modules.
pytorch image-classification self-attention sub-attention attention-bias
Python
•
MIT License
•0•3•0•0•Updated Jun 16, 2023Jun 16, 2023
CEM
Public
EMNLP'22, CEM improves MHCH performance by correcting prediction bias and training an auxiliary cost simulator based on user state and labor cost causal graph, without requiring complex model crafting.
python natural-language-processing tensorflow causal-inference dialogue-systems counterfactual causal-graph
Python
•
MIT License
•0•11•0•0•Updated Oct 9, 2022Oct 9, 2022