Skip to content

Dataset · prquan

STARK spatio-temporal benchmark

prquan/STARK_10k

Benchmark for evaluating spatio-temporal reasoning in LLMs and LRMs.

What's inside

Contains tasks of varying complexity levels for cyber-physical systems such as robotics and autonomous vehicles. Presented as English text data with a volume between 10K and 100K.

What you can do with it

  • Evaluating spatio-temporal reasoning of LLMs and LRMs
  • Testing models for autonomous transport and smart cities
Tasks
Question answering
Languages
English
Size
10K–100K rows
Downloads this month
24.1K
Likes
1

How to load it

from datasets import load_dataset

ds = load_dataset("prquan/STARK_10k")