AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Alethia: a Foundational Encoder for Voice Deepfakes

Yi Zhu, Brahmi Dwivedi, Jayaram Raghuram, Surya Koppisetti

ICML 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Zero-shot robustness and transfer · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add generative flow-matching objective (ablation)Alethia-Base · SDD-Eval-50 (all)EER: +0.5 EER when removedImproved
add generative flow-matching objective (ablation)Alethia-Base · SDD diffusion and flow-matching attacksEER: +5.6 EER when removedImproved
add generative flow-matching objective (ablation)Alethia-Base · SDD other generatorsEER: -0.1 EER when removedNo clear effect
add bottleneck architecture without masking or generative objective (ablation)Alethia-Base · SDD vs WavLM-LargeEER: 0.7% EER improvement, 0.49% accuracy gainImproved
continue pretraining Wav2vec2-Base on 1k hours real+fake speech with masked token predictionW2V-base · SDD 50 datasetsEER: +0.80Hurt
pretrain Wav2vec2-Base from scratch on real+fake speechW2V-base · SDD 50 datasetsEER: +1.20Hurt
pretrain HuBERT-Base from scratch on real+fake speechHuBERT-base · SDD 50 datasetsEER: -0.30No clear effect
continue pretraining HuBERT-Base on real+fake speechHuBERT-base · SDD 50 datasetsEER: +0.25Hurt
raise CNN mask ratio from 10% to 20% (ablation)Alethia-Base · SDD 30 datasetsEER: 16.2 -> 19.2Hurt
raise CNN mask ratio from 10% to 30% (ablation)Alethia-Base · SDD 30 datasetsEER: 16.2 -> 19.3Hurt
use zero mask instead of learnable CNN mask (ablation)Alethia-Base · SDD 30 datasetsEER: 16.2 -> 17.4Hurt
use Gaussian noise mask instead of learnable CNN mask (ablation)Alethia-Base · SDD 30 datasetsEER: 16.2 -> 18.9Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy