Haozhao Wang

BoRA: Towards More Expressive Low-Rank Adaptation with Block Diversity

Breaking Static Paradigms: A Mutual Evolution Framework for Edge-Cloud Model Collaboration

ContrastGen: A Multi-agent Contrastive Framework for Hard Retrieval Data Generation

Enhancing RAG with Active Learning on Conversation Records: Reject Incapables and Answer Capables

You Don't Have to Be the One Doing Evil! Locate a Scapegoat within Multi-Agent Systems for Executing Covert Attacks

SAFD: A Style-Agnostic Framework for Detecting LLM-Generated Text

Model Editing is Over: Revealing Its Illusory Success and Fragile Foundation

C$^2$-AFCL: Cross-task Calibration for Asynchronous Federated Continual Learning

FedSSI: Rehearsal-Free Continual Federated Learning with Synergistic Synaptic Intelligence

Efficient Knowledge Transfer in Federated Recommendation for Joint Venture Ecosystem

Beyond Higher Rank: Token-wise Input-Output Projections for Efficient Low-Rank Adaptation

BSemiFL: Semi-supervised Federated Learning via a Bayesian Approach

ChatbotID: Identifying Chatbots with Granger Causality Test

LLM at Network Edge: A Layer-wise Efficient Federated Fine-tuning Approach

Beyond Zero Initialization: Investigating the Impact of Non-Zero Initialization on LoRA Fine-Tuning Dynamics

Resource-Constrained Federated Continual Learning: What Does Matter?

Feature Distillation is the Better Choice for Model-Heterogeneous Federated Learning

Enhancing Privacy in Multimodal Federated Learning with Information Theory

Self-Introspective Decoding: Alleviating Hallucinations for Large Vision-Language Models

Rehearsal-Free Continual Federated Learning with Synergistic Regularization

Breaking Free from MMI: A New Frontier in Rationalization by Probing Input Utilization

The Panaceas for Improving Low-Rank Decomposition in Communication-Efficient Federated Learning

Adversarial Cooperative Rationalization: The Risk of Spurious Correlations in Even Clean Datasets

Frequency-Decoupled Cross-Modal Knowledge Distillation

Attacking for Inspection and Instruction: Attack Techniques Can Aid In Interpretability