Yiming Wang

Models may ace aesthetics but falter on geometry—SVGEval reveals the critical gaps in text-to-SVG generation evaluation.

Regional bias in LLMs is not just abstract; it significantly influences social decisions and reflects real-world economic disparities.

Pathological markers for Alzheimer's may mislead diagnostic AI, particularly for Greek and male speakers, revealing critical demographic gaps in acoustic biomarker effectiveness.

LLMs struggle significantly with culturally loaded translations, revealing critical gaps in their performance and evaluation methods.

Interleaving speech and text during ASR training boosts entity recognition accuracy and narrows the gap between modalities, challenging traditional training paradigms.

Language-action pretraining can lead to VLA policies that are not only more robust but also less dependent on visual cues, achieving up to 45% higher success rates in real-world tasks.

Quantum pseudorandom states can only be stretched to a limited extent, revealing a stark contrast with classical counterparts.

Encoder-free speech modeling can rival traditional methods, challenging the necessity of dedicated speech encoders in LLM architectures.

ASR-driven data augmentation boosts Alzheimer's detection accuracy by over 4%, showcasing the potential of synthetic speech in clinical diagnostics.

FLAME uncovers a hidden statistical energy gap in AI-generated images, enabling precise localization of forgeries that traditional methods miss.