arXiv cs.MA· Lei Zhenbing, Lei Chang·· 11 小时前AI 评分26
LLM 智能体竞技场 AITRAP 激励机制的实证审计：集中度、奖励规则实现偏差与网络稀释
Incentive Pathologies in a Live LLM Agent Arena: An Empirical Study of Concentration, Reward-Rule Implementation Discrepancy, and Network Dilution
AI 导读
一项对 LLM 智能体竞技场 AITRAP（22 个自主智能体席位、约 96,000 个推理节点）为期九天（10 月 16-25 日）的审计发现，Top-4 节点活跃度得分占比从 97.2%（10 月 19 日）降至 49.4%（10 月 25 日）。
正文 · 原文
Abstract:We audit one live LLM agent arena (AITRAP; 22 autonomous agent seats, ~96,000 reasoning nodes) over nine days (Oct 16-25, 2026), during which the platform deployed two bundled incentive changes: a progressive tax reform (V2.9.3, Oct 19) and a seed-inversion with explorer-side action update (V3.0, Oct 21). These serve as before-and-after markers, not cleanly isolated interventions. Four exploratory predictions are tested against ten data frames (F1-F10) using the Gini coefficient, Top-K concentration share, code-level reconstruction of the continuation-bonus mechanism, and the Mann-Kendall trend test. The Gini analysis is inconclusive at 95% confidence (bootstrap intervals straddle 0.7). The Top-4 node-activity-score share falls from 97.2% (Oct 19, pre-V2.9.3) to 49.4% (Oct 25, post-V3.0; 56.7% under a merge-seat assumption). The continuation bonus, announced as depth-based x10, resolves at code level to a four-condition anti-abuse gate paying x2.85 in the typical case and permanently blocking the self-continuation it superficially rewards; the gate was active ~40 hours before the V3.0 announcement. The events-per-node ratio declines significantly (Mann-Kendall S=-45, z=-3.94, p=8.3e-5) from 7.36 to 7.069. We discuss implications for incentive designers, the announcement-implementation gap, and public peer review as an event record.
| Comments: | 29 pages, 2 figures, 4 tables; data available at Zenodo DOI https://doi.org/10.5281/zenodo.23691548 |
| Subjects: | Multiagent Systems (cs.MA); Physics and Society (physics.soc-ph) |
| Cite as: | arXiv:2611.09445 [cs.MA] |
| (or arXiv:2611.09445v1 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2611.09445 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Lei Zhenbing [view email]
[v1]
Fri, 6 Nov 2026 13:19:48 UTC (388 KB)
来源：arXiv cs.MA · arxiv.org