Skip to content

hadley/dabstep

Repository files navigation

DABstep

The DABstep (Data Agent Benchmark for Multi-Step Reasoning) dataset from Adyen. The dataset contains synthetic payment transaction data along with fee rules, merchant profiles, and 450 benchmark tasks that test multi-step reasoning about payment processing. Includes a data-dictionary.yml that describes what I've learned about the table schemas, relationships, and domain terms.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages