本报告系统梳理了以 foundation model 为核心的代理 AI 系统在多轮执行中的内在随机性来源及其放大效应,强调 token 采样是最直接的随机源,同时外部环境、部署基础设施亦会引入变异,系统区分解码层、模型输出层、代理动作层及业务流程层的变异与后果,便于在实际场景中进行追踪和治理 [page::1]。通过提出在固定权重、确定性解码、以及环境固定等条件下实现可重复性的诊断框架,帮助组织在设计、测试和运营阶段评估变异是否可控、可重现或仅分布式衡量 [page::1]
Multimodal LLMs enable low-cost digitization of heterogeneous historical tables by leveraging domain expertise via prompts, enabling cohesive panel data from diverse sources [page::0]. Using gold-standard data for evaluation, the pipeline achieves 95.4% exact cell matching at roughly 50x lower costs than outsourcing [page::0]. It also dramatically reduces layout parsing errors (61.4% to 0.35%), and attains 96.7% exact transcription of linked cells with a mean absolute percentage error of 0.7% [page::0]. In replication studies, the LLM pipeline aligns with human-based header categorization and yields econometric inferences that are largely indistinguishable from gold-standard results across eight models [page::0].
We estimate Kyle’s price-impact coefficient λ directly from daily equity order flow and test its ability to forecast the cross-section of subsequent stock returns. Using CRSP data from 2020 to 2025, we construct firm-month measures of signed order flow and two estimators of ˆλit: a within-month price-impact regression and an Amihud-style ratio. Signed order flow strongly predicts contemporaneous and one-month-ahead returns, while volume volatility predicts lower subsequent returns, consistent with widening price impact degrading price discovery. Fama-MacBeth regressions confirm that our order-flow signal carries significant cross-sectional return information after Newey–West adjustment. Theoretically, we resolve the liquidity premium puzzle of Constantinides (1986) through an adverse-selection mechanism: low order flow widens λ and depresses prices today; subsequent normalization restores prices, generating the illiquidity premium without risk-based compensation. [page::0]
本研究聚焦趋势跟随在期货市场的长期盈利性及其自2008–2009年后的崩溃现象。以约100份高流动性期货合约的横截面数据和CTA代理为基础,系统比较了容量、电子化、订单流与微观结构等四类解释,最终发现决定趋势生存与否的核心横截面变量是波动率标准化后的跳动尺寸(Ψ/σ)。研究表明:2008年前后,快速信号在小跳合约中的收益迅速下降,而在大跳合约中基本保持;这一现象并不能被资产类别或流动性水平简单解释。作者提出自我强化的价格冲击回路在小跳合约被高频做市商撤出深度供给后被打断,导致趋势信号及其回报显著下降;而大跳合约因仍具充足的市场深度,回路得以维持,趋势依然存在。对价格冲击与执行方式(主动成交 vs 被动成交)的分析也显示,被动执行无法逃离这一回路,因其放弃了自我强化渠道且伴随机会成本。该论文的核心结论强调应以Ψ/σ作为跨品种的容量与风险评估基准,而非简单以资产类别划分。 [page::1]