Topic
zero-shot
ZeSTA Framework Enhances Zero-Shot TTS Augmentation for Data-Efficient Personalized Speech Synthesis
Researchers propose ZeSTA, a domain-conditioned training framework that distinguishes real and synthetic speech via a lightweight domain embedding, combined with real-data oversampling. The approach improves speaker similarity over naive synthetic augmentation while preserving intelligibility and perceptual quality in low-resource personalized speech synthesis.
VERITAS Framework Uses Verifier Feedback to Boost Zero-Shot Theorem Proving Accuracy
VERITAS, a zero-shot framework for formal theorem proving, leverages all verifier signals rather than collapsing them into a binary pass/fail. It reaches 40.6% on the miniF2F benchmark, outperforming Best-of-5 (36.9%) and Portfolio (26.2%). On a new combinatorics benchmark, VERITAS scores 7.3% while unguided sampling falls to 1.8%, demonstrating the value of feedback-driven exploration.
ArtNet: JEPA-Like Articulatory Framework Achieves 20.56% Error Reduction in Zero-Shot Phoneme Recognition
Researchers propose ArtNet, a JEPA-like framework for zero-shot cross-lingual phoneme recognition. By integrating an articulatory predictor with a variational information bottleneck, ArtNet suppresses language-specific variations. Experiments on seven unseen languages show a 20.56% relative reduction in phoneme error rate and 7.01% in phoneme feature error rate.