MasqLoRA shows that an independent LoRA adapter can be trained on a few trigger-target pairs to backdoor diffusion models with 99.8% success rate while remaining stealthy when the trigger is absent.
Laion-5b: An open large-scale dataset for training next generation image-text models.Advances in neural in- formation processing systems, 35:25278–25294
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
fields
cs.CV 6representative citing papers
SteelDefectX is a new multi-form vision-language dataset and benchmark for analyzing steel surface defects using 7,778 images across 25 categories.
O2MAG generates high-fidelity text-guided anomalies from a single image without training by manipulating self-attention in diffusion models with anomaly masks and dual enhancements.
SigLino distills SigLIP2 and DINOv3 into efficient vision models via asymmetric relation-knowledge distillation, token-balanced batching, and hierarchical data sampling on a new 200M-image corpus, yielding better transfer to grounding VLMs than training from scratch.
RADSeg adapts the RADIO model with targeted enhancements to deliver 6-30% higher mIoU in zero-shot OVSS while using 2.5x fewer parameters and running 3.95x faster than prior large-model combinations.
MedBridge adapts pretrained VLMs to multi-label medical diagnosis via query tokens for non-destructive alignment and expert routing, reporting 6-15% AUC gains on chest radiograph benchmarks across eight models.
citing papers explorer
-
When LoRA Betrays: Backdooring Text-to-Image Models by Masquerading as Benign Adapters
MasqLoRA shows that an independent LoRA adapter can be trained on a few trigger-target pairs to backdoor diffusion models with 99.8% success rate while remaining stealthy when the trigger is absent.
-
SteelDefectX: A Multi-Form Vision-Language Dataset and Benchmark for Steel Surface Defect Analysis
SteelDefectX is a new multi-form vision-language dataset and benchmark for analyzing steel surface defects using 7,778 images across 25 categories.
-
One-to-More: High-Fidelity Training-Free Anomaly Generation with Attention Control
O2MAG generates high-fidelity text-guided anomalies from a single image without training by manipulating self-attention in diffusion models with anomaly masks and dual enhancements.
-
SigLino: Efficient Multi-Teacher Distillation for Agglomerative Vision Foundation Models
SigLino distills SigLIP2 and DINOv3 into efficient vision models via asymmetric relation-knowledge distillation, token-balanced batching, and hierarchical data sampling on a new 200M-image corpus, yielding better transfer to grounding VLMs than training from scratch.
-
RADSeg: Unleashing Parameter and Compute Efficient Zero-Shot Open-Vocabulary Segmentation Using Agglomerative Models
RADSeg adapts the RADIO model with targeted enhancements to deliver 6-30% higher mIoU in zero-shot OVSS while using 2.5x fewer parameters and running 3.95x faster than prior large-model combinations.
-
Adapting Foundation Vision-Language Models to Medical Diagnosis via Query-Driven Expert Bridging
MedBridge adapts pretrained VLMs to multi-label medical diagnosis via query tokens for non-destructive alignment and expert routing, reporting 6-15% AUC gains on chest radiograph benchmarks across eight models.