2 papers
cs.CL2026
Early-Token Confidence Predicts Reasoning Quality in Multi-Agent LLM Debate
Ali Keramati, Justin Cheok, Jacob Horne +1
Evaluating reasoning quality in multi-agent LLM systems is challenging, especially for open-ended tasks without reference answers. We investigate whether intrinsic confidence signa…
cs.CL2026
The Confident Liar: Diagnosing Multi-Agent Debate with Log-Probabilities and LLM-as-Judge
Ali Keramati, Justin Cheok, Jacob Horne +1
Multi-agent debate systems are typically evaluated only on whether the final answer is correct, overlooking the quality of the intermediate reasoning that debate is designed to pro…