Picture for Jin Song Dong

Jin Song Dong

On the Mistaken Assumption of Interchangeable Deep Reinforcement Learning Implementations

Add code
Mar 28, 2025
Viaarxiv icon

Verification of Bit-Flip Attacks against Quantized Neural Networks

Add code
Feb 22, 2025
Viaarxiv icon

Enhancing Model Defense Against Jailbreaks with Proactive Safety Reasoning

Add code
Jan 31, 2025
Figure 1 for Enhancing Model Defense Against Jailbreaks with Proactive Safety Reasoning
Figure 2 for Enhancing Model Defense Against Jailbreaks with Proactive Safety Reasoning
Figure 3 for Enhancing Model Defense Against Jailbreaks with Proactive Safety Reasoning
Figure 4 for Enhancing Model Defense Against Jailbreaks with Proactive Safety Reasoning
Viaarxiv icon

Clustering Properties of Self-Supervised Learning

Add code
Jan 30, 2025
Viaarxiv icon

Defending LVLMs Against Vision Attacks through Partial-Perception Supervision

Add code
Dec 17, 2024
Figure 1 for Defending LVLMs Against Vision Attacks through Partial-Perception Supervision
Figure 2 for Defending LVLMs Against Vision Attacks through Partial-Perception Supervision
Figure 3 for Defending LVLMs Against Vision Attacks through Partial-Perception Supervision
Figure 4 for Defending LVLMs Against Vision Attacks through Partial-Perception Supervision
Viaarxiv icon

The Fusion of Large Language Models and Formal Methods for Trustworthy AI Agents: A Roadmap

Add code
Dec 09, 2024
Figure 1 for The Fusion of Large Language Models and Formal Methods for Trustworthy AI Agents: A Roadmap
Figure 2 for The Fusion of Large Language Models and Formal Methods for Trustworthy AI Agents: A Roadmap
Figure 3 for The Fusion of Large Language Models and Formal Methods for Trustworthy AI Agents: A Roadmap
Viaarxiv icon

VisionCoder: Empowering Multi-Agent Auto-Programming for Image Processing with Hybrid LLMs

Add code
Oct 25, 2024
Figure 1 for VisionCoder: Empowering Multi-Agent Auto-Programming for Image Processing with Hybrid LLMs
Figure 2 for VisionCoder: Empowering Multi-Agent Auto-Programming for Image Processing with Hybrid LLMs
Figure 3 for VisionCoder: Empowering Multi-Agent Auto-Programming for Image Processing with Hybrid LLMs
Figure 4 for VisionCoder: Empowering Multi-Agent Auto-Programming for Image Processing with Hybrid LLMs
Viaarxiv icon

Forgetting Through Transforming: Enabling Federated Unlearning via Class-Aware Representation Transformation

Add code
Oct 09, 2024
Figure 1 for Forgetting Through Transforming: Enabling Federated Unlearning via Class-Aware Representation Transformation
Figure 2 for Forgetting Through Transforming: Enabling Federated Unlearning via Class-Aware Representation Transformation
Figure 3 for Forgetting Through Transforming: Enabling Federated Unlearning via Class-Aware Representation Transformation
Figure 4 for Forgetting Through Transforming: Enabling Federated Unlearning via Class-Aware Representation Transformation
Viaarxiv icon

LLM-based Abstraction and Concretization for GUI Test Migration

Add code
Sep 08, 2024
Figure 1 for LLM-based Abstraction and Concretization for GUI Test Migration
Figure 2 for LLM-based Abstraction and Concretization for GUI Test Migration
Figure 3 for LLM-based Abstraction and Concretization for GUI Test Migration
Figure 4 for LLM-based Abstraction and Concretization for GUI Test Migration
Viaarxiv icon

Contribution Evaluation of Heterogeneous Participants in Federated Learning via Prototypical Representations

Add code
Jul 02, 2024
Viaarxiv icon