LMAct Benchmark

What is LMAct?

LMAct is a benchmark for evaluating frontier multimodal models' in-context imitation learning capabilities in long contexts.

Released

2024-12-02

Evaluates

长文本, Long-Context, 智能体, Agent, 任务执行, Task Execution, 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/LMAct

LMAct paper and code

• Paper https://arxiv.org/abs/2412.01441
• Code https://github.com/google-deepmind/lm_act

Which sources cite LMAct?

• LMAct

Related 长文本, Long-Context, 智能体, Agent, 任务执行, Task Execution, 长上下文, Long Context, 不支持, Unsupported benchmarks

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