الباحثون

Xinyuan Qian

المنشورات 2

نسخة أولية وصول مفتوح

SAVU-BENCH: A Real-World Benchmark for Spatial Audio-Visual Understanding

Yu Chen, Ruihang Liu, Yangguang Xu وآخرون · 2026

Spatial audio-visual understanding requires models to recognize not only what is present, but also where events occur and how they relate across modalities. Existing benchmarks often rely on simulated scenes, evaluate isolated spatial skills, and provide limited diagnostic insight into failure modes. We introduce SAVU- …

نسخة أولية وصول مفتوح

FragToken: Amplifying LLM Inference Costs through Noncanonical Token Generation

Zihan Wang, Rui Zhang, Xinyuan Qian وآخرون · 2026

As large language model (LLM) inference becomes increasingly expensive, resource-consumption attacks pose a growing threat to model providers. Existing attacks typically amplify cost by inducing abnormally long or repetitive outputs on attacker-controlled or triggered requests, making them easier to detect and limiting …

المؤلفون المشاركون

Yu Chen Ruihang Liu Yangguang Xu Xinyue Jiang Mohammed Bennamoun Farid Boussaid Qiuhong Ke Zihan Wang Rui Zhang Qingchuan Zhao Hong-Wei Li Guowen Xu