arXiv cs.CL· Zeyu Liu, Sing Kwong Lai, Xuanzhi Feng, HaoYang Shang, Yang Li, Zihao Fang, Xuan Liu·· 11 小时前AI 评分38
AStockArena：超越收益率，能否把真实资金托付给金融智能体的决策？
AStockArena: Beyond Returns, Can We Entrust Real Capital to the Decisions of Financial Agents?
AI 导读
AStockArena 是面向中国 A 股连续交易的实时决策审计基准，在相同市场条件与执行约束下评估了 13 个开源与闭源模型的 3,198 次连续决策。在 1,070 个交易点上，72.99% 至少存在一项审计缺口，模型层面缺口率为 52.2% 至 91.2%，其中证据不足最为普遍。历史交易过滤回放显示组合结果对这些被标记交易敏感，决策过程审计可作为收益类评估的可验证补充。
正文 · 原文
Abstract:Large language models (LLMs) are increasingly used in dynamic financial trading, yet real-time trading benchmarks mainly evaluate agents by cumulative returns and risk-adjusted performance. Such outcome metrics cannot show whether trades are faithfully reported, supported by available evidence, and consistent with account states. We introduce AStockArena, a real-time decision-auditing benchmark for continuous trading in China's A-share market. Under shared market conditions and execution constraints, we evaluate 13 open- and closed-source models over 3,198 sequential decisions. AStockArena reconstructsthe actions actually posted to accounts by aligning reports, trade requests, execution responses, and account states, and audits them with eight checks spanning execution traceability (O), evidence sufficiency (RK), and financial meta-awareness (B), combining record verification with consensus expert annotation. Of 1,070 trading points, 72.99% exhibit at least one audit gap; model-level rates range from52.2% to 91.2%, with evidence gaps most prevalent. Historical trade-filtering replay further shows thatportfolio outcomes are sensitive to these flagged trades. Decision-process auditing thus offers a verifiable complement to return-based evaluation of financial agents.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.00515 [cs.CL] |
| (or arXiv:2611.00515v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.00515 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: HaoYang Shang [view email]
[v1]
Sat, 31 Oct 2026 01:08:02 UTC (388 KB)
来源：arXiv cs.CL · arxiv.org