Back to paper
Full reasoning
Sep 11, 2026
cs.AI · cs.CL
Share this scorecard
Every score comes with the model's reasoning on the paper page.
Suggested post
"Do LLMs Trust the Accuser or the Accusation? Measuring Belief Shifts in Werewolf" scored 5.5/10 predicted impact on @kurateorg — Clarity 8.0 · Reproducible 8.0 · Rigor 6.5