GRASP applies deterministic conditioning-space partitioning and sample-wise residual adapters to improve tail-class fidelity, diversity, and downstream utility in flow matching models, outperforming full fine-tuning and MoE baselines on medical and ImageNet long-tail data.
Dinov2: Learning robust visual features with- out supervision
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2verdicts
UNVERDICTED 2representative citing papers
BlendFusion uses path tracing on 3D scenes with targeted camera placement to produce higher-quality synthetic image-caption data for diffusion model training than direct generation methods.
citing papers explorer
-
GRASP: Guided Residual Adapters with Sample-wise Partitioning
GRASP applies deterministic conditioning-space partitioning and sample-wise residual adapters to improve tail-class fidelity, diversity, and downstream utility in flow matching models, outperforming full fine-tuning and MoE baselines on medical and ImageNet long-tail data.
-
BlendFusion -- Scalable Synthetic Data Generation for Diffusion Model Training
BlendFusion uses path tracing on 3D scenes with targeted camera placement to produce higher-quality synthetic image-caption data for diffusion model training than direct generation methods.