TonyL

TonyMou

·

TonyMou

AI & ML interests

None yet

Organizations

None yet

Collections 2

models 0

None public yet

datasets 0

None public yet

TonyMou (TonyL)

Hugging Face

TonyL

TonyMou

·

TonyMou

AI & ML interests

None yet

Organizations

None yet

Collections 2

models 0

None public yet

datasets 0

None public yet

TonyL

AI & ML interests

Organizations

Collections 2

EfficientQAT: Efficient Quantization-Aware Training for Large Language Models

PrefixQuant: Static Quantization Beats Dynamic through Prefixed Outliers in LLMs

OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models

An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models

TokenPacker: Efficient Visual Projector for Multimodal LLM

DiffRate : Differentiable Compression Rate for Efficient Vision Transformers

EfficientQAT: Efficient Quantization-Aware Training for Large Language Models

PrefixQuant: Static Quantization Beats Dynamic through Prefixed Outliers in LLMs

OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models

An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models

TokenPacker: Efficient Visual Projector for Multimodal LLM

DiffRate : Differentiable Compression Rate for Efficient Vision Transformers

models 0

datasets 0

TonyL

AI & ML interests

Organizations

Collections 2

EfficientQAT: Efficient Quantization-Aware Training for Large Language Models

PrefixQuant: Static Quantization Beats Dynamic through Prefixed Outliers in LLMs

OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models

An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models

TokenPacker: Efficient Visual Projector for Multimodal LLM

DiffRate : Differentiable Compression Rate for Efficient Vision Transformers

EfficientQAT: Efficient Quantization-Aware Training for Large Language Models

PrefixQuant: Static Quantization Beats Dynamic through Prefixed Outliers in LLMs

OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models

An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models

TokenPacker: Efficient Visual Projector for Multimodal LLM

DiffRate : Differentiable Compression Rate for Efficient Vision Transformers

models 0

datasets 0