5 papers
When EER Hides Deployment Failure: Auditing Threshold Transfer and Unlabeled Score Calibration for Speech Deepfake Detectors
Jingwen Zhou, Mingzhe Wang
Speech deepfake countermeasures (CMs) are compared almost exclusively by equal error rate (EER), a metric computed at an oracle threshold chosen on the labeled test set. Deployed C…
How Fragile Are Training-Free AI-Generated Image Detectors? A Controlled Audit of Score Direction, Preprocessing, and Compression
Jingwen Zhou, Mingzhe Wang
Training-free detectors of AI-generated images promise generator-agnostic deployment without classifier training, yet their reported numbers are rarely compared under a single cont…
Scoring Backends Matter More Than Pooling: A Systematic Study of Training-Free Anomalous Sound Detection under Domain Shift
Jingwen Zhou, Mingzhe Wang
Training-free anomalous sound detection (ASD) scores a test clip against a memory bank of normal embeddings from a frozen pretrained audio encoder. Recent work attributes domain-sh…
Transductive Zero-Shot Audio Classification with Audio-Language Models
Jingwen Zhou, Mingzhe Wang
Contrastive language-audio pretraining (CLAP) enables zero-shot audio classification, but standard inference classifies each clip in isolation and ignores the structure of the unla…
False Sense of Safety in Selective Signal Classification: Auditing Bound Tightness and Exchangeability for Risk Control
Jingwen Zhou, Mingzhe Wang
Selective prediction with distribution-free risk control promises that, with confidence 1-delta over the calibration draw, the error rate of accepted inputs stays below a user budg…