The paper proposes a transparent, user‑configurable rule for selecting arguments in deliberative polls, replacing opaque learned rankers. It formalises argument selection over bipolar justification sets, introduces seven civic recommender criteria, and presents a one‑hop reversed endorsement flow rule that meets them. Experiments on 17,000 simulated runs show the rule performs comparably to random on coverage but outperforms other methods on endorsement mass and robustness under adversarial pressure.
arXiv:2606. 11692v1 Announce Type: cross Abstract: Deliberative polling promises to improve collective decision-making by exposing shareholders to a broad range of arguments before they vote.
arXiv:2606. 30931v1 Announce Type: new Abstract: The LLM Jury, a Panel of LLM Evaluators (PoLL) reporting consensus scores, has become a practical alternative to single-judge LLM evaluation, yet its statistical behavior remains poorly understood.
The paper investigates proportional representation in a temporal voting setting where a single candidate is chosen each round and voters submit ranked preferences that may evolve over time. It extends classic proportionality axioms—justified representation (JR), proportional JR (PJR), extended JR (EJR), and proportionality for solid coalitions (PSC)—to accommodate various ways of determining which top-ranked candidates are considered approved, ranging from a fixed common cutoff to individual, round‑specific cutoffs. The authors analyze which axioms can be guaranteed under different informational assumptions about future rounds, showing that while EJR is unattainable, JR, PJR, and PSC can be achieved with a fixed cutoff if all preferences are known in advance; varying cutoffs reduce guarantees, yet PJR can still be achieved efficiently for groups that agree in every round, and PSC can be satisfied without future knowledge. They also demonstrate that checking these axioms is often coNP‑complete, though some stronger axioms may be easier to verify.
The paper studies how to safely delegate action approval to multiple AI reviewers when the reviewers themselves may be misaligned. It introduces a weaker condition—k‑robust coalitional alignment—under which a threshold rule that tolerates up to k disapprovals guarantees that the principal’s expected utility is at least as good as a baseline policy. The authors extend this characterization to sequential decision‑making in discounted MDPs and show that full‑panel coverage of reward functions ensures safety in Nash equilibria, while more permissive thresholds can lead to unsafe outcomes. Experiments demonstrate that collective review can remain sound even when individual reviewers are not fully aligned, provided some disapprovals are allowed.
arXiv:2606. 08098v1 Announce Type: new Abstract: Majority voting over sampled answers is the dominant unsupervised aggregator for multi-sample LLM inference.
arXiv:2606. 00914v1 Announce Type: new Abstract: LLM agents increasingly act after consuming ranked external information streams such as social feeds, search results, retrieval contexts, and email queues, yet safety evaluations almost always test the model or the user prompt in isolation, never the upstream ranker that decides what the agent reads just before it acts.
JuryProbe is an empirical diagnostic tool designed to assess consensus risk in panels of reference‑free large language model judges used for factuality verification. It estimates risk by measuring false‑negative correlations and false‑consensus lift from a labeled calibration probe, and routes high‑risk majority decisions to judges with trusted references. The approach was validated on FEVER corruptions, showing that flagged decisions can be grounded without additional reference acquisition in most cases, while reducing false accepts by about 0.4% and avoiding 28% of reference acquisitions.
arXiv:2606. 07834v1 Announce Type: cross Abstract: LLM judges increasingly turn verdicts into system commitments.
arXiv:2608. 16813v1 Announce Type: new Abstract: Agents now write knowledge graphs, but knowledge-graph stores still carry defaults set when humans curated them: accept writes now and clean later, keep one time axis or none, treat every writer's facts as equally trustworthy, and leave governance to dashboards and middleware.
arXiv:2608. 06422v1 Announce Type: new Abstract: Giving an LLM judge more compute does not necessarily make it check more requirements.
arXiv:2503. 01985v2 Announce Type: replace-cross Abstract: In the classic committee election setting each voter approves a subset of candidates and the goal is to select $k$ winners based on these preferences.
arXiv:2607. 18960v1 Announce Type: cross Abstract: Procuring supervised fine-tuning (SFT) data forces a buyer to decide, before any downstream training, whether a candidate corpus is worth acquiring.