arXiv:2607. 22034v1 Announce Type: cross Abstract: Vision-language models (VLMs) are increasingly deployed on consumer hardware where input images are degraded by compression, camera shake, and poor lighting.
arXiv:2609.00868v1 Announce Type: cross Abstract: Vision-language models are evaluated by aggregate accuracy on multimodal benchmarks, a practice that implicitly assumes the model uses its visual inp...
SpatialUQ is a post‑hoc uncertainty quantification method that uses only the output probabilities of frozen black‑box vision models. It computes the Jensen‑Shannon divergence between the global prediction and the mean of five fixed spatial crops across six deterministic forward passes, assuming that trustworthy predictions are spatially consistent. On the NIH ChestX‑ray14 dataset, SpatialUQ’s Multicrop Uncertainty Score (MUS) achieves a 0.784 failure‑detection AUC—significantly higher than MC‑Dropout—while also providing superior native calibration and improving further when fused with other uncertainty signals.
arXiv:2610.19763v1 Announce Type: new Abstract: Physics-guided networks for underwater image restoration predict transmission and backscatter and invert the image-formation model, yet rarely test whe...
arXiv:2609.09184v1 Announce Type: new Abstract: Vision-language model (VLM) confidence may change in aggregate when visual evidence is degraded while remaining structurally inconsistent within indivi...
arXiv:2606. 15910v2 Announce Type: replace Abstract: A vision-language model can answer a question about a chest radiograph or a pathology slide fluently and confidently while barely using the image, relying instead on language priors.
arXiv:2609. 11916v1 Announce Type: new Abstract: Camera traps often run in the field on edge hardware with limited or no connectivity, making small, locally-deployable vision-language models (VLMs) -- not frontier-scale ones -- the practically relevant class to evaluate for species identification.
arXiv:2606.15910v3 Announce Type: replace Abstract: An answer from a vision-language model (VLM) to a medical image question can be accepted without review only under a threshold bounding the error o...
The paper introduces SphereTrust, a method that uses frozen self‑supervised hyperspherical features to evaluate and rank candidate masks produced by foundation segmenters like SAM. By measuring angular contrast, foreground coverage, and image‑frame contact, SphereTrust can select high‑quality masks in 0.55 s per image and outperforms existing baselines on multiple segmentation tasks. The selected masks are then used as priors to train student models, improving performance on several benchmark datasets.
arXiv:2608.29705v1 Announce Type: cross Abstract: Feed-forward 3D reconstruction models emit a per-pixel confidence that downstream systems read as a reliability signal. It is trained as a loss weigh...
arXiv:2608.29193v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) can assign similar confidence to answers that fail for different reasons. We propose HalluPrism, a behavioral...
arXiv:2610.28000v1 Announce Type: cross Abstract: VLMs for medical image interpretation are typically evaluated on clean inputs, despite the progressive evidence loss that can occur in high stakes cl...