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

arXiv: https://arxiv.org/abs/2606.10908 · HTML: https://mlsanity.com/paper/2606.10908/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 31 references that are in the map.
- First posted on arXiv 9 Jun 2026.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://mlsanity.com/paper/2111.09296/) (2021)
- [AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks](https://mlsanity.com/paper/2110.01200/) (2021)
- [ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale](https://mlsanity.com/paper/2408.08739/) (2024)
- [RawBoost: A Raw Data Boosting and Augmentation Method applied to Automatic Speaker Verification Anti-Spoofing](https://mlsanity.com/paper/2111.04433/) (2021)
- [Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals](https://mlsanity.com/paper/2007.05979/) (2020)
- [SASV 2022: The First Spoofing-Aware Speaker Verification Challenge](https://mlsanity.com/paper/2203.14732/) (2022)
- [Graph Attention Networks for Anti-Spoofing](https://mlsanity.com/paper/2104.03654/) (2021)
- [BUT Systems and Analyses for the ASVspoof 5 Challenge](https://mlsanity.com/paper/2408.11152/) (2024)
- [Differential Anomaly Detection for Facial Images](https://mlsanity.com/paper/2110.03464/) (2021)
- [Malafide: a novel adversarial convolutive noise attack against deepfake and spoofing detection systems](https://mlsanity.com/paper/2306.07655/) (2023)
- [Exploring WavLM Back-ends for Speech Spoofing and Deepfake Detection](https://mlsanity.com/paper/2409.05032/) (2024)
- [USTC-KXDIGIT System Description for ASVspoof5 Challenge](https://mlsanity.com/paper/2409.01695/) (2024)
- [Diffuse or Confuse: A Diffusion Deepfake Speech Dataset](https://mlsanity.com/paper/2410.06796/) (2024)
- [Malacopula: adversarial automatic speaker verification attacks using a neural-based generalised Hammerstein model](https://mlsanity.com/paper/2408.09300/) (2024)
- [Learn from Real: Reality Defender's Submission to ASVspoof5 Challenge](https://mlsanity.com/paper/2410.07379/) (2024)
- [SCDF: A Speaker Characteristics DeepFake Speech Dataset for Bias Analysis](https://mlsanity.com/paper/2508.07944/) (2025)
- [SZU-AFS Antispoofing System for the ASVspoof 5 Challenge](https://mlsanity.com/paper/2408.09933/) (2024)
- [Revisiting and Improving Scoring Fusion for Spoofing-aware Speaker Verification Using Compositional Data Analysis](https://mlsanity.com/paper/2406.10836/) (2024)
- [Evolutionary Multi-Objective Fusion of Deepfake Speech Detectors](https://mlsanity.com/paper/2604.01330/) (2026)
- [Hybrid Pruning: In-Situ Compression of Self-Supervised Speech Models for Speaker Verification and Anti-Spoofing](https://mlsanity.com/paper/2508.16232/) (2025)
- [Speaker-Aware Anti-Spoofing](https://mlsanity.com/paper/2303.01126/) (2023)

## Abstract (arXiv)

We introduce a spoofing countermeasure architecture conditioned on speaker-reference recordings, but observe that it converges to a solution that effectively ignores the reference during inference. Surprisingly, training with a reference channel induces invariance that improves deepfake detection, even when the reference is absent or mismatched during inference. Based on this observation, we propose a Reference-Augmented Training (RAT) strategy. RAT yields improved detection performance compared to single-utterance baselines, even when the reference recording is replaced with a zero vector at inference. Through rigorous analysis, we demonstrate that the optimization process rapidly diminishes the reference contributions, leading to inference largely independent of the reference channel. Using RAT, we achieve state-of-the-art 2.57% EER and 0.074 minDCF on the ASVspoof 5 benchmark with a single detector, surpassing even large ensemble systems.