Yue Yu

Preserve and Sculpt: Manifold-Aligned Fine-tuning of Vision-Language Models for Few-Shot Learning

SecP-Tuning: Efficient Privacy-Preserving Prompt Tuning for Large Language Models via MPC

GEPO: Group Expectation Policy Optimization for Stable Heterogeneous Reinforcement Learning

Map as a Prompt: Learning Multi-Modal Spatial-Signal Foundation Models for Cross-scenario Wireless Localization

Self-Guidance: Training VQ-VAE Decoders to be Robust to Quantization Artifacts for High-Fidelity Neural Speech Codec

Meta-Memory for Large Language Models

Simple Convergence Proof of Adam From a Sign-like Descent Perspective

SpecEM: Training-Free LLM Ensembling via Iterative Drafting, Verification, and Online Feedback

Centrality-guided Pre-training for Graph

SoftSignSGD(S3): An Enhanced Optimize for Practical DNN Training and Loss Spikes Minimization Beyond Adam

SpecFuse: Ensembling Large Language Models via Next-Segment Prediction

Cross-modal Mitigation of Spurious Correlation for Prompt-tuning in VLMs with Causally Motivated Logic Alignment