Reproduce “Inter-Agent Influence: Evaluating Persuasion, Deception, and Coercion in Multi-Agent Systems” 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

Chandler Smith, Qi Guo, Cecilia Elena Tilli, Sophia Hatz, David Demitri Africa, Patricia Paskov, Christian Schroeder de Witt, Philip Torr and 1 more

The authors evaluate whether language-model agents can influence others through persuasion, deception, or coercion in five settings. They compare influence prompts with baselines in oversight and peer tasks. Deception raises approval probability by 31.20 percentage points in oversight results, while persuasion raises research pivots by 21.36 points; these prompt effects do not isolate techniques or establish effects beyond the tested settings.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.