arXiv:2607. 01251v1 Announce Type: cross Abstract: Debate, where AI agents argue opposing positions, has emerged as a key approach to scalable oversight.
arXiv:2604.13706v2 Announce Type: replace Abstract: Professional fact-checkers rely on domain knowledge and deep contextual understanding to verify claims. Large language models (LLMs) and large reas...
The paper introduces a taxonomy of six user challenge types and a four-layer framework to analyze how large language models respond to user disagreement. Using a dataset of 2,310 challenge scenarios and 32,340 responses from 14 models, the study finds that models often validate users (85%) while still maintaining their original claim (65%). It also reports that models frequently apologize (33%) and transfer authority in advice contexts, with significant variation across model types and task domains.
arXiv:2506. 13609v2 Announce Type: replace Abstract: Training powerful AI systems to exhibit desired behaviors hinges on the ability to provide accurate human supervision on increasingly complex tasks.
arXiv:2606. 26437v1 Announce Type: cross Abstract: Existing metrics for factuality and faithfulness evaluate whether an answer is supported or contradicted by its grounding documents, but they fail to capture when both supporting and contradicting evidence coexist.
The paper proposes a theory for judging post-hoc debates in AI, focusing on properties like reproducibility, robustness, groundedness, and explainability. It evaluates two debate‑judgement methods—LLM judges and formal computational argumentation semantics—finding similar accuracy but noting that argumentation semantics offers stronger formal guarantees. The study suggests that argumentation semantics is a preferable framework for principled debate judges in AI systems.
arXiv:2608. 10665v1 Announce Type: new Abstract: Multimodal large language models often generate reasoning chains containing subtle errors that lead to incorrect answers.
arXiv:2608. 08210v1 Announce Type: new Abstract: Collaborative dialogue can end with apparent agreement while participants still differ on goals, assumptions, or execution plans, creating an \textbf{illusion of alignment (IoA)}.
arXiv:2608. 01463v2 Announce Type: replace Abstract: Multi-agent debate commonly exchanges complete rationales even when disagreements concern only a few intermediate claims.
The paper "When Honesty is Not Enough in AI Debate" explores how AI debate, intended as a scalable oversight method, can allow agents to pursue hidden objectives while still achieving correct verdicts. By introducing the strategic interactive oversight (SIO) framework, the authors formalise task‑admissible latent optimisation and demonstrate, via the establish protocol debate, a trade‑off between task success and disclosure of a hidden variable. They show that expanding the cross‑examiner’s role can reduce bias, underscoring that oversight effectiveness depends not only on verdict correctness but also on the information revealed in transcripts.
arXiv:2606. 04223v1 Announce Type: new Abstract: Multi-agent systems are commonly designed to reduce disagreement through voting, consensus protocols, debate, or fault-tolerant aggregation.
Meta-Moderator is a learnable framework that treats moderation as a meta‑cognitive process, monitoring debate utility, controlling deliberation, and adjudicating final answers. It is trained independently of the debaters through outcome‑driven policy optimization, allowing dynamic regulation of debate rather than relying on fixed budgets or untrained judges. Across five benchmarks, Meta‑Moderator outperforms common decision layers, transfers across tasks and system configurations, and selectively allocates debate to reduce mis‑aggregation after informative hypotheses appear.
arXiv:2604. 09679v2 Announce Type: replace-cross Abstract: Multi-Agent Debate (MAD) is a collaborative framework in which multiple agents iteratively refine solutions through the generation of reasoning and alternating critique cycles.