Two separate context-dependent memory selectors that learn to identify useful experience for the current trading decision are developed that enable the Trader to improve its proposals based on prior critiques and the Red-Team to adjust its review focus as the Trader's behavior changes.
Abstract
Generator--critic trading systems pair an LLM-based trader that makes trading decisions with a red team agent that evaluates these decisions and provides feedback for revision. However, improving individual trading decisions does not ensure that both agents learn from experience and improve together over time. This raises an important question: how can the trader and red-Team select useful experience and learn from each other's evolving behavior? On one hand, experience that benefits decisions under one market condition may become ineffective or harmful under another. On the other hand, trading outcomes reflect both agents'actions, making it difficult to derive appropriate experience for each role. In this work, we propose ACE-Trader, an adversarial co-evolution trading framework. Specifically, we first develop separate context-dependent memory selectors that learn to identify useful experience for the current trading decision. Both selectors receive shared feedback from net-return differences between complete memory-assisted and memory-free trading pipelines, with online updates emphasizing recent outcomes. We further design a co-evolution mechanism that links role-specific memories through shared trading records. We use these records to evaluate the Red-Team's risk assessments and the effects of the Trader's revisions against realized market outcomes, then derive experience for their respective memories. By linking this experience to the corresponding records, we enable the Trader to improve its proposals based on prior critiques and the Red-Team to adjust its review focus as the Trader's behavior changes. Experiments on U.S. stock trading show that ACE-Trader outperforms all 22 active baselines in cumulative return and achieves the highest Sharpe ratio and lowest maximum drawdown across both active methods and passive benchmarks.
Multi-agent trading systems built on large language models (LLMs) are beginning to appear in quantitative finance, yet their robustness to adversarial inputs is largely unknown. We study the vulnerability of LLM trading stacks to black-box, input-only attacks that enter solely via admissible social-media feeds. We intr...
Sua-Qi Rong, Jun-Hao Dong, T. Nguyen et al.· ECML PKDD· 1 citation
Experiments show that EvolveTrade often improves Sharpe Ratio and Cumulative Return over fixed-policy LLM baselines, achieving the improved SR and CR in most evaluated settings, and suggest that adapting the reusable procedure governing tool use is a key direction for building more robust LLM trading agents.
Sehee Kim, Yumin Choi, Minki Kang et al.· arXiv.org· 1 citation
A hybrid trading framework that combines a Rule-Based Trading Strategy with a Deep Reinforcement Learning (DRL) agent for intraday gold trading on a 15-minute timeframe and compares the three fixed-RR models to determine which risk–reward configuration provides the most favorable balance between profitability, risk, an...
Ravi Prajapati, Kobid Karkee· Proceedings of International...· 0 citations
A unified benchmark that evaluates representative machine learning, reinforcement learning, LLM-based, and agent-based trading methods in cryptocurrency markets through three progressively more realistic stages through historical backtesting, prospective exchange-based paper trading, and real-money live trading is pres...
This work introduces LiveOption, an evaluation framework for LLM-based agents in option trading that formulates the problem as structured sequential decision-making under realistic execution and capital constraints, and provides a reproducible environment with standardized interaction protocols.
Hao-Chen Luo, Yi-Fan Li, An Binh Minh et al.· arXiv.org· 0 citations
Controlled evidence ablations show that recent trading history strongly governs activity prediction, whereas asset selection is substantially more sensitive to the available evidence.