JavisGPT: A Unified Multi-modal LLM for Sounding-Video Comprehension and Generation
Zhejiang University · The Hong Kong University of Science and Technology (GZ) · Renmin University of China · National University of Singapore · University of Rochester · Nanyang Technological University · Singapore Management University · University of Sydney, University of Sydney · Xiamen University · Australian National University · National University of Singapore, Department of Electrical and Computer Engineering · National Univ. of Singapore
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。
摘要
This paper presents JavisGPT, the first unified multimodal large language model (MLLM) for Joint Audio-Video (JAV) comprehension and generation. JavisGPT adopts a concise encoder–LLM–decoder architecture, featuring a SyncFusion module for spatio-temporal audio- video fusion and synchrony-aware learnable queries to bridge a pretrained JAV-DiT generator. This design enables temporally coherent video-audio understanding and generation from multimodal instructions. We design an effective three-stage training pipeline consisting of multimodal pretraining, audio-video fine-tuning, and large-scale instruction-tuning, to progressively build multimodal comprehension and generation from existing vision-language models. To support this, we further construct JavisInst-Omni, a high-quality instruction dataset with over 200K GPT-4o-curated audio-video-text dialogues that span diverse and multi-level comprehension and generation scenarios. Extensive experiments on JAV comprehension and generation benchmarks show that JavisGPT outperforms existing MLLMs, particularly in complex and temporally synchronized settings.