Xinyuan Song

Kuromi: Learning without Augmentation via Energy-based Semi-supervised Kuramoto Neurons

What Governs the Quality-Aware Generalization and Representation Capacity in No-Reference Image Quality Assessment Models?

CML-Bench: A Framework for Evaluating and Enhancing LLM-Powered Movie Scripts Generation

GPAS: Accelerating Convergence of LLM Pretraining via Gradient-Preserving Activation Scaling

The Curse of Depth in Large Language Models

TRiCo: Triadic Game-Theoretic Co-Training for Robust Semi-Supervised Learning