Zheng Wu

This paper provides the first comprehensive security analysis of the Agent Skills framework, identifying severe structural vulnerabilities that require fundamental architectural changes rather than simple mitigations.

EigeNet introduces a geometry-informed multi-modal Transformer framework to achieve state-of-the-art few-shot novel view Room Impulse Response (RIR) prediction by effectively integrating spatial geometry and multi-view acoustic context.

The paper introduces MineExplorer, a new benchmark in Minecraft, to evaluate the sustained open-world exploration capabilities of MLLM agents, finding that long-horizon coordination remains a significant challenge.

This paper introduces Speaker Anonymization (SA) as a novel perturbation mechanism for zero-shot voice conversion, balancing timbre leakage and prosodic utility while enabling strictly causal, zero-lookahead networks.

This study investigates defensive driving behaviors, their impact on driving safety, and underlying mechanisms through driving simulator experiments.

This paper introduces a training recipe for sentence-level and long-form streaming speech-to-speech translation using only 2k hours of paired cross-lingual data and auxiliary supervision.