Paper: arXiv 2610.30557

Authors: Yebo Feng, Jingyang Dai, Liangrui Pan, Nelson Lau, Jiahua Xu, Yang Liu

Abstract

LLM agents that act on live Web data must combine heterogeneous evidence, remember what they have done and improve from outcomes without weakening their own safeguards. We study this problem in autonomous investing, where every action is irreversible and based on untrusted Web content. We present Steward, a multi-agent LLM system that draws the boundary of self-evolution in its type system. First, nine role-specialised agents decide jointly: four analysts read news feeds, SEC filings, macroeconomic series and market data and emit typed reports, from which a Portfolio Manager sets target positions that a protected Risk Manager audits. Second, a seven-layer shared memory, assembled by code from the system’s own record, gives each decision its position context, the market narrative, calibration and the agent’s past decisions. Third, the system improves itself by review: code grades every trade and forecast nightly and writes the grades into memory, and each quarter a reflector proposes prompt learnings that a deterministic gate applies reversibly and replay of recorded inputs verifies. All three operate around an invariant core of order arithmetic, risk rules and discipline-agent prompts that no self-edit can reach. In a five-month live deployment (102 trading days, 1,632 LLM calls), the system contained a 24.6% sector drawdown to a 0.3% loss; in replay of its recorded inputs, a single model given the same raw data traded more and lost 1.35 points against the market over five days where the joint design showed no significant loss, and no injected instruction was followed; and a 2015-2026 backtest shows the core alone lowering the maximum drawdown under all 50 random entry policies. The reflector traced the dominant failure to its root cause, and the gate applied a learning that the agent adopted on trigger evenings and on no control evening. Code: https://github.com/yebof/quant-agent

Complexity vs Empirical Score

  • Math Complexity: 5.0/10
  • Empirical Rigor: 6.0/10 (capped from 9.0: transaction costs not modelled; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: This paper presents a highly novel multi-agent LLM system for autonomous investing with a strong emphasis on governed self-evolution and safety. Its empirical rigor is exceptional, featuring a five-month live deployment and a comprehensive backtest, while also providing open-source code for reproducibility. The blend of advanced AI concepts with real-world financial application makes it a significant contribution.

Research Flowchart

  flowchart TD
    A[Research Goal: Autonomous Investing with Self-Evolving LLM Agents & Safeguards] --> B{Methodology: Multi-Agent LLM System (Steward)};
    B --> C[Data/Inputs: Live Web Data (News, SEC, Macro, Market)];
    C --> D{Computational Processes: <br>1. Role-Specialized Agents (Analysts, PM, Risk Manager)<br>2. 7-Layer Shared Memory<br>3. Self-Improvement by Review (Grading, Reflector, Gate)};
    D --> E[Invariant Core: Order Arithmetic, Risk Rules, Discipline Prompts];
    E --> F[Key Findings/Outcomes: <br>- Contained 24.6% drawdown to 0.3% loss in live deployment <br>- Outperformed single model baseline <br>- Core lowered max drawdown in backtest <br>- Reflector identified failure, Gate applied learning];