Read the Paper, Write the Code: Agentic Reproduction of Social-Science Results
This work addresses the problem of automating the reproduction of social science research for researchers and reviewers, but the results are incremental, showing feasibility with high variability.
The paper investigates whether LLM agents can reproduce empirical social science results given only a paper's methods description and original data, without seeing the original code or results. Evaluating four agent scaffolds and four LLMs on 48 papers, they find that agents can largely recover published results, but performance varies substantially between models, scaffolds, and papers.
Recent work has used LLM agents to reproduce empirical social science results with access to both the data and code. We broaden this scope by asking: Can they reproduce results given only a paper's methods description and original data? We develop an agentic reproduction system that extracts structured methods descriptions from papers, runs reimplementations under strict information isolation -- agents never see the original code, results, or paper -- and enables deterministic, cell-level comparison of reproduced outputs to the original results. An error attribution step traces discrepancies through the system chain to identify root causes. Evaluating four agent scaffolds and four LLMs on 48 papers with human-verified reproducibility, we find that agents can largely recover published results, but performance varies substantially between models, scaffolds, and papers. Root cause analysis reveals that failures stem both from agent errors and from underspecification in the papers themselves.