Delegated Influence

A competitive multi-agent benchmark for LLM persuasion: the only way to score is to get other agents to spend their scarce actions on you.

287958d · generated 2026-07-03 · 40 episodes · private draft — not for citation

Experiment

Big arena

Experiment 12 — big arena: the full roster in one game (descriptive; Q5 emergence). Descriptive only: produces standings, coalition structure, and cascade counts; no model claims.

status
planned
coverage
0 / 6 episodes
conditions
complete+ring topology; pure economy; messages on; seeds 41,42,43
questions
q5
config
configs/12_arena.yaml

Planned — 6 episodes from 12_arena.yaml.

image/svg+xml Matplotlib v3.11.0, https://matplotlib.org/ P L A N N E D 6 episodes · 12_arena.yaml

Episodes

No episodes yet — launch with:

uv run python -m delegated_influence.run configs/12_arena.yaml