Jul 2026
MultiRef-Compass: Towards Comprehensive Evaluation of Multi-Reference-to-Audio-Video Generation
MultiRef-Compass is introduced, a unified benchmark for MR2AV generation that integrates automatic metrics with a rejudging-enhanced MLLM-as-a-Judge framework, enabling scalable and auditable evaluation of both perceptual fidelity and reference-conditioned composition.
Xiaohan Zhang, Yu-Qing Wen, Jun-Lin Chen et al.
· arXiv.org · 3 citations