7 papers
VoxZip: Semantic-Anchored Temporal KV Cache Compression for Long-Context Audio Inference
Wenxu Jia, Dongjie Fu, Xize Cheng +6
Recent advancements in Speech Large Language Models have demonstrated remarkable capabilities in understanding complex audio tasks. Despite this progress, their long-context infere…
X-OPD: Distilling Reasoning into Large Audio-Language Models via On-Policy Alignment
Dongjie Fu, Di Cao, Xize Cheng +6
While large audio-language models have achieved remarkable progress in auditory perception, they still lag behind text-based large language models in deep logical reasoning, primar…
Auto-AEG: Scalable Data Construction for Open-Vocabulary Audio Event Grounding
Zihan Zhang, Xize Cheng, Wenhao Yan +5
Large Audio-Language Models (LALMs) reason fluently about sound yet struggle to localize precisely when events occur, while classical Sound Event Detection attains frame-level prec…
X-OPD: Cross-Modal On-Policy Distillation for Capability Alignment in Speech LLMs
Di Cao, Dongjie Fu, Hai Yu +3
While the shift from cascaded dialogue systems to end-to-end (E2E) speech Large Language Models (LLMs) improves latency and paralinguistic modeling, E2E models often exhibit a sign…
Character Beyond Speech: Leveraging Role-Playing Evaluation in Audio Large Language Models via Reinforcement Learning
Dongjie Fu, Fangming Feng, Xize Cheng +3
The rapid evolution of multimodal large models has revolutionized the simulation of diverse characters in speech dialogue systems, enabling a novel interactive paradigm. Character…
MARS-Sep: Multimodal-Aligned Reinforced Sound Separation
Zihan Zhang, Xize Cheng, Zhennan Jiang +4
Universal sound separation faces a fundamental misalignment: models optimized for low-level signal metrics often produce semantically contaminated outputs, failing to suppress perc…