Benchmark

LongVALE Benchmark

What is LongVALE?

LongVALE is the first long video benchmark integrating fine-grained omni modalities (video, audio, speech) information in videos. It comprises 105K omni-modal…

Released: 2025-04-17
Modality: multimodal
Evaluates: 多模态, Multimodal, 理解, Understanding, 长文本, Long-Context, long video understanding, omni-modal, fine-grained temporal understanding, 多模态模型, VLM, 视频理解, Video Understanding, 长上下文, Long Context, 不支持, Unsupported
Openness: unknown
Importer: OpenCompass Hub
Reported scores: 0

Source provenance

Original evidence https://hub.opencompass.org.cn/dataset-detail/LongVALE

LongVALE paper, code and dataset

Paper https://arxiv.org/abs/2411.19772
Code https://github.com/ttgeng233/LongVALE
Dataset https://huggingface.co/datasets/ttgeng233/LongVALE

No reported scores are on record for this benchmark yet.

Which sources cite LongVALE?

LongVALE

Related 多模态, Multimodal, 理解, Understanding, 长文本, Long-Context, long video understanding, omni-modal, fine-grained temporal understanding, 多模态模型, VLM, 视频理解, Video Understanding, 长上下文, Long Context, 不支持, Unsupported benchmarks