R2-AVSBench
收藏资源简介:
R2-AVSBench是一个新的评估基准,设计用于包含更多具有语言多样性和推理密集性的参考。它旨在更好地评估模型在跨引用场景中的性能。该数据集由 Mohamed Bin Zayed University of Artificial Intelligence, National University of Singapore, University of Science and Technology of China 研究机构的研究人员创建,但论文中并未提供数据集的具体条数和访问地址。
R2-AVSBench is a novel evaluation benchmark designed to incorporate more linguistically diverse and reasoning-intensive reference instances. It aims to better evaluate the performance of models in cross-referencing scenarios. This dataset was developed by researchers from Mohamed Bin Zayed University of Artificial Intelligence, National University of Singapore, and University of Science and Technology of China. However, the specific number of samples and the access URL of the dataset are not provided in the associated paper.

- 1Think Before You Segment: An Object-aware Reasoning Agent for Referring Audio-Visual Segmentation Mohamed Bin Zayed University of Artificial Intelligence, National University of Singapore, University of Science and Technology of China · 2025年



