ShiZhengyan/StepGame

[AAAI 2022] Dataset and pytorch codes for the paper titled "StepGame: A New Benchmark for Robust Multi-Hop Spatial Reasoning in Texts" in AAAI 2022 (Oral)

39
/ 100
Emerging

This project offers a specialized dataset and accompanying code for evaluating how well AI models can understand and reason about spatial relationships described across multiple sentences in text. It takes text-based stories and questions about spatial arrangements as input, and outputs labels indicating correct spatial reasoning. This is primarily useful for AI researchers and natural language processing engineers who are developing and testing advanced AI systems.

No commits in the last 6 months.

Use this if you are developing or benchmarking AI models that need to perform complex spatial reasoning across several pieces of information within a text.

Not ideal if you are looking for a pre-trained model to directly apply to a business problem, as this is a research benchmark.

AI model evaluation Natural Language Processing research textual reasoning spatial understanding AI benchmark
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

32

Forks

7

Language

Python

License

MIT

Last pushed

Mar 20, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/ShiZhengyan/StepGame"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.