Lattice AI Research

Research focus

Speech & Audio (3)Red-Teaming & Adversarial Robustness (2)Interpretability & Mechanistic Interp (1)RLHF & Preference Learning (1)

Frequent co-authors

Junichi Yamagishi (2)Viola Negroni (1)Paolo Bestagini (1)P. Bestagini (1)

Papers (3)

Jul 23, 2026

Jul 23, 2026·also NII

Toward Interpretable Speech Deepfake Detection using Artifact-Specific Experts and Calibrated Detection Scores

Artifact-specific experts can provide interpretable evidence in speech deepfake detection, transforming how we assess synthetic audio authenticity.

Viola Negroni, Wanying Ge, Paolo Bestagini +3

Interpretability & Mechanistic Interp Speech & Audio

Mar 3, 2026

Mar 3, 2026·also NII

Does Fine-tuning by Reinforcement Learning Improve Generalization in Binary Speech Deepfake Detection?

RL fine-tuning with GRPO can significantly boost the generalization of speech deepfake detectors to unseen attacks, outperforming standard supervised methods.

Xin Wang, Ge Wanying, Wanying Ge +1

Red-Teaming & Adversarial Robustness RLHF & Preference Learning Speech & Audio

Feb 26, 2026

Feb 26, 2026·also Edinburgh

Deepfake Word Detection by Next-token Prediction using Fine-tuned Whisper

Repurpose Whisper as a deepfake detector and you can transcribe and spot synthetic words in one pass, rivaling dedicated models.

Hoan My Tran, Hoan My Tran, Wanying Ge +1

Natural Language Processing Red-Teaming & Adversarial Robustness Speech & Audio

Search

Wanying Ge

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (3)