↓Skip to main content

🏢 Southern University of Science and Technology

Time-Varying LoRA: Towards Effective Cross-Domain Fine-Tuning of Diffusion Models
·3031 words·15 mins· loading · loading
Computer Vision Image Generation 🏢 Southern University of Science and Technology
Terra, a novel time-varying low-rank adapter, enables effective cross-domain fine-tuning of diffusion models by creating a continuous parameter manifold, facilitating efficient knowledge sharing and g…
SeTAR: Out-of-Distribution Detection with Selective Low-Rank Approximation
·5697 words·27 mins· loading · loading
AI Generated Multimodal Learning Vision-Language Models 🏢 Southern University of Science and Technology
SeTAR: Training-free OOD detection via selective low-rank approximation, improving robustness and efficiency.
Does Egalitarian Fairness Lead to Instability? The Fairness Bounds in Stable Federated Learning Under Altruistic Behaviors
·1528 words·8 mins· loading · loading
Machine Learning Federated Learning 🏢 Southern University of Science and Technology
Achieving egalitarian fairness in federated learning without sacrificing stability is possible; this paper derives optimal fairness bounds considering clients’ altruism and network topology.
Boundary Decomposition for Nadir Objective Vector Estimation
·2892 words·14 mins· loading · loading
AI Theory Optimization 🏢 Southern University of Science and Technology
BDNE: a novel boundary decomposition method accurately estimates the nadir objective vector in complex multi-objective optimization problems.
Association of Objects May Engender Stereotypes: Mitigating Association-Engendered Stereotypes in Text-to-Image Generation
·3516 words·17 mins· loading · loading
Image Generation 🏢 Southern University of Science and Technology
New framework, MAS, effectively mitigates stereotypes in text-to-image generation by aligning the probability distribution of generated images to stereotype-free distributions.