Picture for Mingjie Sun

Mingjie Sun

ConSense: Continually Sensing Human Activity with WiFi via Growing and Picking

Add code
Feb 18, 2025
Viaarxiv icon

Idiosyncrasies in Large Language Models

Add code
Feb 17, 2025
Viaarxiv icon

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Add code
Feb 03, 2025
Viaarxiv icon

SweetTokenizer: Semantic-Aware Spatial-Temporal Tokenizer for Compact Visual Discretization

Add code
Dec 17, 2024
Figure 1 for SweetTokenizer: Semantic-Aware Spatial-Temporal Tokenizer for Compact Visual Discretization
Figure 2 for SweetTokenizer: Semantic-Aware Spatial-Temporal Tokenizer for Compact Visual Discretization
Figure 3 for SweetTokenizer: Semantic-Aware Spatial-Temporal Tokenizer for Compact Visual Discretization
Figure 4 for SweetTokenizer: Semantic-Aware Spatial-Temporal Tokenizer for Compact Visual Discretization
Viaarxiv icon

Bi-Mamba: Towards Accurate 1-Bit State Space Models

Add code
Nov 18, 2024
Viaarxiv icon

FBI-LLM: Scaling Up Fully Binarized LLMs from Scratch via Autoregressive Distillation

Add code
Jul 09, 2024
Figure 1 for FBI-LLM: Scaling Up Fully Binarized LLMs from Scratch via Autoregressive Distillation
Figure 2 for FBI-LLM: Scaling Up Fully Binarized LLMs from Scratch via Autoregressive Distillation
Figure 3 for FBI-LLM: Scaling Up Fully Binarized LLMs from Scratch via Autoregressive Distillation
Figure 4 for FBI-LLM: Scaling Up Fully Binarized LLMs from Scratch via Autoregressive Distillation
Viaarxiv icon

Massive Activations in Large Language Models

Add code
Feb 27, 2024
Figure 1 for Massive Activations in Large Language Models
Figure 2 for Massive Activations in Large Language Models
Figure 3 for Massive Activations in Large Language Models
Figure 4 for Massive Activations in Large Language Models
Viaarxiv icon

Reliable Test-Time Adaptation via Agreement-on-the-Line

Add code
Oct 07, 2023
Viaarxiv icon

Dance with You: The Diversity Controllable Dancer Generation via Diffusion Models

Add code
Sep 04, 2023
Viaarxiv icon

A Simple and Effective Pruning Approach for Large Language Models

Add code
Jun 20, 2023
Viaarxiv icon