BABILong Benchmark

What is BABILong?

BABILong is designed to test language models' ability to reason across facts distributed in extremely long documents. It contains a diverse set of 20 reasoning…

Released

2024-06-14

Evaluates

长文本, Long-Context, NeurIPS 2024, 大语言模型, LLM, 长上下文, Long Context, 开源收录, Open-Source, 不支持, Unsupported

Openness

restricted

Importer

OpenCompass Hub

Reported scores

0

Source provenance

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

BABILong paper, code and dataset

• Paper https://arxiv.org/abs/1502.05698
• Paper https://arxiv.org/abs/2406.10149
• Paper https://identifiers.org/arxiv:2402.10790
• Code https://github.com/booydar/babilong
• Dataset https://huggingface.co/datasets/RMT-team/babilong
• Dataset https://huggingface.co/datasets/facebook/babi_qa

Which sources cite BABILong?

• BABILong

Related 长文本, Long-Context, NeurIPS 2024, 大语言模型, LLM, 长上下文, Long Context, 开源收录, Open-Source, 不支持, Unsupported benchmarks

• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• Ada-LEval
• AECBench
• AFQMC
• AgentBoard