Picture for Xiaodong Liu

Xiaodong Liu

BayLing 2: A Multilingual Large Language Model with Efficient Language Alignment

Add code
Nov 25, 2024
Viaarxiv icon

Adaptively Controllable Diffusion Model for Efficient Conditional Image Generation

Add code
Nov 19, 2024
Viaarxiv icon

Map-Free Trajectory Prediction with Map Distillation and Hierarchical Encoding

Add code
Nov 17, 2024
Viaarxiv icon

StreamAdapter: Efficient Test Time Adaptation from Contextual Streams

Add code
Nov 14, 2024
Viaarxiv icon

Model Editing for LLMs4Code: How Far are We?

Add code
Nov 11, 2024
Viaarxiv icon

Generative Adapter: Contextualizing Language Models in Parameters with A Single Forward Pass

Add code
Nov 08, 2024
Figure 1 for Generative Adapter: Contextualizing Language Models in Parameters with A Single Forward Pass
Figure 2 for Generative Adapter: Contextualizing Language Models in Parameters with A Single Forward Pass
Figure 3 for Generative Adapter: Contextualizing Language Models in Parameters with A Single Forward Pass
Figure 4 for Generative Adapter: Contextualizing Language Models in Parameters with A Single Forward Pass
Viaarxiv icon

Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities

Add code
Oct 24, 2024
Figure 1 for Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities
Figure 2 for Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities
Figure 3 for Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities
Figure 4 for Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities
Viaarxiv icon

GRIN: GRadient-INformed MoE

Add code
Sep 18, 2024
Figure 1 for GRIN: GRadient-INformed MoE
Figure 2 for GRIN: GRadient-INformed MoE
Figure 3 for GRIN: GRadient-INformed MoE
Figure 4 for GRIN: GRadient-INformed MoE
Viaarxiv icon

Model Tells Itself Where to Attend: Faithfulness Meets Automatic Attention Steering

Add code
Sep 16, 2024
Figure 1 for Model Tells Itself Where to Attend: Faithfulness Meets Automatic Attention Steering
Figure 2 for Model Tells Itself Where to Attend: Faithfulness Meets Automatic Attention Steering
Figure 3 for Model Tells Itself Where to Attend: Faithfulness Meets Automatic Attention Steering
Figure 4 for Model Tells Itself Where to Attend: Faithfulness Meets Automatic Attention Steering
Viaarxiv icon

Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning

Add code
Aug 26, 2024
Figure 1 for Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning
Figure 2 for Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning
Figure 3 for Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning
Figure 4 for Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning
Viaarxiv icon