Inside the Trading Loop: Markets, Models, and Moves with LLM Agents
Abstract
A hybrid multi-agent trading system combining large language model analysts with momentum-based portfolio construction and rule-based execution produced a 34.5% paper-account return from March 11 to August 28, 2026, exceeding SPY's 14.4% but with a larger maximum drawdown.
We built a hybrid multi-agent trading system that combines large language model (LLM) analysts with momentum-based portfolio construction and rule-based execution. Company-news and macroeconomic agents return structured reports to a coordinator that sets stock weights and cash reserves. We evaluate the portfolio engine using its broker-recorded paper account, which operated with deterministic fallbacks, and compare LLM-enabled configurations through retrospective replays. From March 11 to August 28, 2026, the account returned 34.5%, compared with 14.4% for the S&P 500 fund SPY over the same dates. This raw return advantage came with a larger maximum drawdown. Reconstructed policies also finish above matched SPY and QQQ references when assessments are used in the same session or deferred to the next. Design comparisons examine cash authority, selective invocation, and response contracts. A constant-cash control gives a return estimate close to stance-based sizing, and shorter report contracts improve observed parsing but can change recommendations.
Community
🚀 Can LLM Agents Really Trade? We Put Them Inside the Trading Loop.
What happens when AI agents go beyond analyzing markets—and start influencing real portfolio decisions?
In Inside the Trading Loop, we explore the intersection of multi-agent intelligence, quantitative finance, and autonomous decision-making through an end-to-end trading architecture connecting market analysis, portfolio allocation, and execution.
📈 34.5% vs. 14.4% SPY: Our broker-recorded paper-trading portfolio delivered a 34.5% return over the study period, compared with 14.4% for SPY, albeit with substantially higher drawdown. The recorded account used deterministic fallbacks, while LLM-enabled strategies were examined through retrospective replays.
🔬 Beyond returns, we investigate a deeper question: When does LLM intelligence actually translate into better decisions?
We explore:
- Agent-to-Action: How structured LLM insights influence portfolio weights and capital allocation.
- Intelligence on Demand: When selective agent invocation can preserve decision impact while reducing unnecessary computation.
- The Reliability Paradox: Why making LLM outputs easier to parse can unexpectedly change their recommendations.
- From Reasoning to Execution: How agent authority, risk exposure, and stateful execution
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- How Execution Assumptions Change Short-Horizon Sharpe Rankings: Evidence from a Synthetic Trading Benchmark (2026)
- When News Gets Old: Portfolio Allocation with Reinforcement Learning and Large Language Models (2026)
- The Price of Thought: Does Test-Time Reasoning Pay in LLM Trading? (2026)
- FIDES-Live: Evaluating Behavioral Concordance in Sequential Trading Agents (2026)
- Foresee to Trade: Forecast-Augmented Policy Optimization for Portfolio Management (2026)
- Quant-Arena: Benchmarking LLM Agents for Quantitative Trading Research (2026)
- When Is Fresh AI Stock Analysis Worth It? Allocating a Limited Research Budget Across Price Updates, Deep Forecast Refreshes, and Editorial Reviews of Top Stock Picks (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend
Get this paper in your agent:
hf papers read 2611.11657 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper