Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.33728

Strategic Calibration in LLM Multi-Agent Systems: Audited Pivot Scoring for Truthful Elicitation

Jinhao Zhou·Zhaohan Zhang·Xijian Xu·Jun Wu
View original source ↗Read source document ↗Export metadata ↓

Abstract

In large language model (LLM)-based multi-agent systems, reported confidence increasingly determines the allocation of scarce downstream attention, computation, and decision influence. This makes reporting strategic: an agent may overstate the relevance of its contribution even when its underlying belief is well calibrated. Existing calibration methods largely assume passive reporting and therefore do not address strategic incentives created when confidence determines downstream allocation. We formulate strategic calibration as a reporting-layer incentive problem and propose the Audited Pivot Scoring Mechanism (APSM). APSM offsets platform-controlled allocation benefits, uses inverse-propensity-weighted sparse audits, charges the Clarke pivot externality, and applies a strictly proper score. Under the stated model assumptions, truthful reporting is the unique expected-utility best response. We also characterize approximate execution and the audit-variance trade-off. Calculator-free experiments show that LLM agents inflate confidence under Top-$K$ incentives. With a shared agent-local utility interface, LLM agents reduce distortion under both Compensated Proper Scoring (CPS) and APSM. CPS achieves higher exact allocation match, whereas APSM yields lower Brier loss and numerically fewer cutoff crossings. Experiments further show that APSM enlarges the utility margin against allocation-changing deviations by $1.93\times$ and improves robustness under approximate execution. These results support incentive-aware elicitation as a distinct layer for reliable multi-agent coordination.

Explore related subjects

cs.GTcs.AI

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Jinhao Zhou, Zhaohan Zhang, Xijian Xu, Jun Wu. 2026-10-24. Strategic Calibration in LLM Multi-Agent Systems: Audited Pivot Scoring for Truthful Elicitation. https://arxiv.org/abs/2610.33728

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Prophet Inequalities with Delayed and Uncertain Acceptance

We introduce the *prophet inequality with delayed and uncertain acceptance* model, a variant of the classical prophet inequality in which a decision-maker sequentially evaluates options whose acceptance is uncertain and revealed only after a random delay. At each time step, the decision-maker observes the realized value of the arriving option and must irrevocably decide whether to select it or continue searching. Once an option is selected, the process is suspended for a delay drawn from a known distribution with expectation $d$, during which no other option can be considered. When the delay expires, the selection succeeds with a known probability. If it succeeds, the decision-maker receives the realized value and the process terminates; otherwise, the search resumes. In addition to the online decision-maker, we consider two stronger benchmarks: the value-aware decision-maker, who knows all value realizations in advance but not the acceptance outcomes, and the prophet, who knows both the values and the acceptance outcomes. We show that, in the worst case, both decision-makers achieve a $1/(d + 2)$ fraction of the prophet's value, while the decision-maker achieves half of the value-aware decision-maker's value. We then turn to the no-delay case, where we establish an equivalence between our model and the classical prophet inequality, and show that the value-aware decision-maker beats the $1/2$ barrier against the prophet whenever all acceptance probabilities are bounded away from $0$.

cs.GT↗
arXiv2026

Position Auctions with a Capacity Constraint

Sponsored search auctions are commonly modeled as an assignment of a fixed set of positions to a set of advertisers, with welfare maximization reducible to a standard matching problem. Motivated by modern advertising environments, we study a richer variant of the classical position auctions model, in which ads may have heterogeneous sizes and the platform must jointly select a subset of ads and assign them to positions subject to a global capacity constraint. The size of an ad can capture the physical space it occupies on a page or its consumption of a scarce platform resource, such as communication bandwidth or computational capacity. Thus, size measures the resource cost imposed by an ad on the platform, which need not be related to the value generated by displaying it. We begin by studying a general formulation of the matching with a capacity constraint problem and propose an algorithmic technique that goes beyond known greedy methods while achieving constant-factor approximation guarantees for social welfare. Our allocation rule augments density-based ordering with capacity-aware local improvements, which allow for re-allocations that improve welfare, while respecting the capacity constraint. Applied in the context of position auctions, we analyze this algorithm under the standard assumptions of single-parameter agents, publicly known ad sizes, and position-dependent click-through rates (CTRs). In this setting, the size of an ad determines the amount of capacity it consumes, while the CTR depends only on the position to which it is assigned. We show that a minor modification to our algorithmic approach yields a universally truthful mechanism that achieves a 4-approximation of the optimal social welfare, and we show that this guarantee is tight. To the best of our knowledge, this is the first such mechanism for this variant of the capacity-constrained matching.

cs.GT↗
arXiv2026

Otter: A Provably MEV-Resilient Automated Market Maker via Surplus Redistribution

Miner extractable value (MEV) in automated market makers allows block builders to profit from transaction ordering and injected trades, imposing costs on users and contributing to builder centralization. We introduce Otter (Optimal Truthful Trading with Excess Redistribution), a two-asset batch AMM that achieves provable MEV resilience when the consensus layer provides censorship resilience and block space is uncongested. Otter makes truthful behavior a dominant strategy for both users and builders. Consequently, a builder cannot profit from strategic deviations, including reordering bids or injecting sybil bids, even when the builder is itself a trader with intrinsic value. Moreover, we show that Otter maximizes social welfare, in a strong sense, within a natural class of mechanisms satisfying the desired game-theoretic properties. To achieve these guarantees, we introduce a new paradigm called surplus redistribution, which provably prevents residual surplus from being captured as MEV by redirecting it to the broader community. Specifically, the pool's output tokens need not be distributed entirely among users in the current batch. Instead, residual surplus may be transferred to a smart contract governed by the decentralized community. The accumulated surplus can be used to benefit community members --- for example, by subsidizing transaction fees, rewarding liquidity providers, or returning assets to the pool. Our approach relies on the underlying consensus layer to provide censorship resilience. We motivate its necessity through an impossibility result showing that the desired game-theoretic guarantees become unattainable when the builder is additionally allowed to censor transactions. Thus, our results also provide a mathematically formal demonstration of how consensus-level security guarantees can fundamentally expand what is achievable at the application layer.

cs.GT↗