Entity · paper

RAT: Reference-Augmented Training for ASV Anti-Spoofing

paperactiverat-reference-augmented-training-for-asv-anti-spoofing-5663735f·1 events·first seen Jun 10, 2026

Aliases: RAT: Reference-Augmented Training for ASV Anti-Spoofing

Co-occurring entities

More like this (12)

ASVspoof 5 Reference-Augmented Training From Self-Supervised Speech Models to Mixture-of-Experts for Robust Anti-Spoofing Automated Reference Verification System ART (Agent Reinforcement Trainer)Automatic Speech Recognition Quantisation-Aware Training (QAT)Verifier-in-the-Loop Training (ViL)APS-RAG Selective Ground Truth Token Training Voice Activity Detection (VAD)AudioVAE2

Recent events (1)

5arXiv · cs.AI·Jun 10, 2026·source ↗

RAT: Reference-Augmented Training improves deepfake audio detection without reference at inference

Researchers introduce Reference-Augmented Training (RAT), a training strategy for automatic speaker verification (ASV) anti-spoofing that conditions a model on speaker-reference recordings during training but discovers the model learns to ignore the reference at inference. Counterintuitively, this training regime induces invariances that improve deepfake detection even when the reference is replaced with a zero vector at test time. RAT achieves state-of-the-art 2.57% EER and 0.074 minDCF on the ASVspoof 5 benchmark with a single detector, outperforming large ensemble systems.

Evaluation and Benchmarking AI Safety Research RAT: Reference-Augmented Training for ASV Anti-Spoofing Reference-Augmented Training ASVspoof 5