让大模型在不知股票名的情况下选股，验证其真懂市场而非靠记忆
Can Blindfolded LLMs Still Trade? An Anonymization-First Framework for Portfolio Optimization
- 用匿名化屏蔽股票代码和公司名，逼大模型靠逻辑推理判断投资价值
- 2025年前8月平均夏普比率达1.40，20次随机种子测试结果稳定
- 适用于验证金融AI是否真具备泛化能力，尤其适合关注模型可信度的研究者
为了让大语言模型交易代理真正值得信赖，必须证明它们理解市场动态，而非依赖预训练中记住的股票代码关联。构建负责任的多智能体系统需要严格的信号验证：确保预测反映真实规律，而非预训练记忆。我们解决两个虚假表现来源：基于股票代码的预训练记忆偏差，以及回测中因幸存者偏差导致的问题。方法是让代理‘戴眼罩’——对所有标识符进行匿名化，并验证有意义的信号是否仍能持续。BlindTrade 匿名化股票代码与公司名称，四个大模型生成评分并附推理过程。我们基于推理嵌入构建图神经网络，采用 PPO-DSR 策略进行交易。在2025年至今（截至2025-08-01）的测试中，20个种子下夏普比率达1.40 ± 0.22，通过负向控制实验验证了信号合法性。为评估跨市场周期的鲁棒性，进一步扩展至2024–2025年期，发现该策略在波动剧烈时期表现优异，但在单边上涨的牛市中收益降低。
原文摘要 · Abstract (English)
For LLM trading agents to be genuinely trustworthy, they must demonstrate understanding of market dynamics rather than exploitation of memorized ticker associations. Building responsible multi-agent systems demands rigorous signal validation: proving that predictions reflect legitimate patterns, not pre-trained recall. We address two sources of spurious performance: memorization bias from ticker-specific pre-training, and survivorship bias from flawed backtesting. Our approach is to blindfold the agents--anonymizing all identifiers--and verify whether meaningful signals persist. BlindTrade anonymizes tickers and company names, and four LLM agents output scores along with reasoning. We construct a GNN graph from reasoning embeddings and trade using PPO-DSR policy. On 2025 YTD (through 2025-08-01), we achieved Sharpe 1.40 +/- 0.22 across 20 seeds and validated signal legitimacy through negative control experiments. To assess robustness beyond a single OOS window, we additionally evaluate an extended period (2024--2025), revealing market-regime dependency: the policy excels in volatile conditions but shows reduced alpha in trending bull markets.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。