Reproduce “Synthetic Conversations for Social Engineering Defense: A Survey of Generation, Validity, and Evaluation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Sanjay Doppalapudi
The authors survey synthetic conversations for scam-defense research, classifying studies by data source, generation, roles, available evidence, and outcomes. Six checks assess label validity, fidelity, behavior, defensive transfer, independence, and reproducibility; they are not a validated score. On Mirage’s held-out real contact-center task, real-transcript prompts had mean macro-F1 0.664 versus 0.598 for the generated-transcript condition, not a scam-detection result.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.