Xian-Ling Mao

Papers

Total Citations

H-Index

About

Xian-Ling Mao is a leading researcher in multimodal perception and artificial intelligence, with a primary focus on advancing audio-visual semantic segmentation (AVSS) and related cross-modal understanding. His most notable contribution is the development of ALOHA (Adapting Local Spatio-Temporal Context), a pioneering framework that fundamentally rethinks how machines integrate audio and visual information for pixel-level scene understanding. Unlike prior approaches that rely on global spatio-temporal fusion, Mao’s work demonstrates that leveraging local spatio-temporal context significantly enhances the alignment and segmentation of audio-visual data, a breakthrough critical for real-world applications such as robotic navigation and autonomous driving. This work, published in 2025, has already garnered attention with 2 citations, reflecting its early impact in a rapidly evolving field. Mao’s research addresses a core challenge in multimodal perception: enabling systems to precisely associate sounds with visual objects in dynamic environments. By introducing a paradigm shift from global to local context adaptation, he has opened new avenues for more efficient and accurate AVSS models. His contributions are poised to influence future developments in embodied AI, where robust audio-visual understanding is essential for safe and intelligent autonomous systems.