Yao Zhu

Generation then Reconstruction: Accelerating Masked Autoregressive Models via Two-Stage Sampling

Pruning as a Cooperative Game: Surrogate-Assisted Layer Contribution Estimation for Large Language Models

Enhancing Reasoning in Large Language Models via Entropy-Aware Self-Evolution

Transferable and Stealthy Adversarial Attacks on Large Vision-Language Models

PlugGuard: A Streaming Safeguard for Large Models via Latent Dynamics-Guided Risk Detection

Mitigating Error Propagation in Low-Rank Approximation of Large Models via Distribution-Aware Whitening

SOPSeg: Prompt-based Small Object Instance Segmentation in Remote Sensing Imagery

Boosting Domain Generalization in Object Detection through the Lens of Phase Invariance

FlattenGPT: Depth Compression for Transformer with Layer Flattening

Adaptive Nonlinear Compression for Large Foundation Models

MemOracle: Symbolic Reasoning with Associative Retrieval for Harmful Content Detection

FlyLoRA: Boosting Task Decoupling and Parameter Efficiency via Implicit Rank-Wise Mixture-of-Experts