Zayne Sprague et al.
Testing the Limits of Chain-of-thought with Multistep Soft Reasoning
A dataset for evaluating language models on multistep soft reasoning tasks specified in natural language narratives. Tests the ability to perform complex, structured reasoning.
ModelsModels with a result.6
Top resultBest result on this benchmark.—
Top-3 spreadPoints from first to third.—
YearYear of release.2023
Results
6 results#ModelResultThe score from the source.