New preprint: CausalDS

Last updated on Jul 9, 2026 1 min read

We recently submitted a new preprint titled CausalDS: Benchmarking Causal Reasoning in Data-Science Agents to ArXiv.

CausalDS is a benchmark generator for causal reasoning in agentic data-science workflows: every benchmark instance is fully synthetically generated, and tests the agents along five separate axes of agentic causal data science.

Joint work with Yuekai Sun.

Causal Inference Large Language Models Agents Benchmarks Data Science Statistics
Andrej Leban
Andrej Leban
Ph.D. Student