Xin Xu

Audio and acceleration modalities can predict physical interactions with surprising accuracy, but the executor's architecture is the true bottleneck in achieving effective multimodal execution.

Heavy speaker overlap drastically hinders speech recognition accuracy in smart glasses, revealing critical limitations in current audio-language models.

Mechanist uncovers a surprising safety risk where unsafe traits can transfer across modalities, challenging assumptions about training data safety.

Disjunctive information flow policies can be enforced with a type system that reveals surprising limitations in traditional methods, leading to stricter precision guarantees.

ACTS achieves full-thinking performance with up to 40% fewer tokens, enabling precise control over reasoning efficiency and accuracy.

Existing music editing systems often compromise on preserving key musical elements, but MuseCPEval provides a robust framework to ensure these facets remain intact during editing.