arXiv AI
Common-Witness Certificates and Sharp Feature Bounds for Counterfactual Image Auditing
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Computer Vision
1d ago
SafeRestore: Detector-Relative Risk Certificates for Selective Industrial Image Restoration
SafeRestore introduces a framework for certifying when an industrial image restoration should be automatically returned to a detector or require human review. It ranks five restoration candidates using action‑specific fitted scores, selects a threshold gate on tuning data, and evaluates the gate on a separate certification sample with two one‑sided exact binomial bounds—one for evidence‑loss incidents and one for excess‑activation incidents. In a retrospective study of 4,591 Carinthia‑S images, the protocol demonstrates auditable risk‑coverage behavior, with varying pass rates across different policies and morphologies.
arXiv Computer Vision
Aug 24
Counterfactual Stress Testing for Image Classification Models
arXiv:2605.10894v2 Announce Type: replace Abstract: Deep learning models in medical imaging often fail when deployed in new clinical environments due to distribution shifts in demographics, scanner h...
arXiv AI
Jun 10
A Controlled Audit of Pretraining Contamination in Public Medical Vision-Language Benchmarks
arXiv:2606. 10066v1 Announce Type: cross Abstract: Medical vision-language models (VLMs) are evaluated on public benchmarks whose images and question-answer pairs have been freely downloadable for years, yet reported accuracy assumes these examples were absent from pretraining.
arXiv Computer Vision
1d ago
Hold-Out Self-Validation Cannot Certify Photogrammetric Accuracy: Saturation and Blindness to Coherent Distortion
arXiv:2607.24852v4 Announce Type: replace Abstract: Internal self-consistency cannot certify the accuracy of a photogrammetric reconstruction, and the failure is structural rather than a matter of tu...
arXiv Computer Vision
4d ago
RePair: Turning Retrieval Failures into Counterfactual Hard Pairs
arXiv:2608.29604v1 Announce Type: cross Abstract: Vision-language retrieval with CLIP-style dual encoders achieves strong cross-modal performance, yet practical accuracy often hinges on localized sem...
arXiv AI
2d ago
Not All Agreement Counts as Corroboration: Provenance-Conserving Multi-View Fusion for Typed Action Admission in Human-Robot Collaboration
The paper introduces PACT, a provenance‑conserving fusion method for typed action admission in human‑robot collaboration. PACT treats evidence countability as a relational variable, using a supplied provenance partition to define countable units and accumulating support only across these units. Experiments on 31,200 evaluations in 48 scene clusters show that PACT achieves a lower normalized risk‑coverage area than singleton aggregation, and in offline human‑robot collaboration it admits 47 of 57 reference‑consistent candidates without any reference‑inconsistent admissions.