Yushun Dong

Reforming the Mechanism: Editing Reasoning Patterns in LLMs with Circuit Reshaping

TRIDENT: An Efficient Data-Free Model Extraction Attack for Graph Neural Networks

CREDIT: Certified Defense of Deep Neural Networks against Model Extraction Attacks

TIFO: Time-Invariant Frequency Operator for Stationarity-Aware Representation Learning in Time Series

CITED: A Decision Boundary-Aware Signature for GNNs Towards Model Extraction Defense

EEG-RAGNet: Retrieval-Augmented Graph Structure Refinement for Clinical Seizure Diagnosis

Tracing LLM Reasoning Processes with Strategic Games: A Framework for Resource-Constrained Decision Making and Revision

Cold Start in the Dark: Efficient and Practical Model Extraction of GNNs

COMPOL: A Unified Neural Operator Framework for Scalable Multi-Physics Simulations

Revisiting Multivariate Time Series Forecasting with Missing Values

GraphIP–Bench: A Benchmark for Extraction Attacks and Defenses in Graph Learning

SOMA: Efficient Multi-turn LLM Serving via Small Language Model

MISLEADER: Defending against Model Extraction with Ensembles of Distilled Models

DESIGN: Encrypted GNN Inference via Server-Side Input Graph Pruning

PCA Feature Alignment is Sufficient for Building Graph Foundation Models

CEB: Compositional Evaluation Benchmark for Fairness in Large Language Models

Certified Defense on the Fairness of Graph Neural Networks

Hierarchical Demonstration Order Optimization for Many-shot In-Context Learning

CEGA: A Cost-Effective Approach for Graph-Based Model Extraction and Acquisition

Towards Global-level Mechanistic Interpretability: A Perspective of Modular Circuits of Large Language Models

Graph Neural Networks Are More Than Filters: Revisiting and Benchmarking from A Spectral Perspective

Fairness-Aware Graph Learning: A Benchmark

Hierarchical Demonstration Order Optimization for Many-shot In-Context Learning

FredNormer: Frequency Domain Normalization for Non-stationary Time Series Forecasting