Traders detect mismatches in sequential trading decisions and reports
FIDES-Live: Evaluating Behavioral Concordance in Sequential Trading Agents
Cryptography and SecurityArtificial Intelligence
Summary
Trading systems make many linked decisions and reports that should match up, but sometimes they don’t. The authors created FIDES-Live, a tool that checks if a trading agent’s different steps like decisions, orders, and reports agree with each other. They used it on thousands of cryptocurrency trades and a small stock trading test. The tool found some mismatches and shows that evaluating various parts like decisions and reports separately helps understand where problems happen, but it doesn’t prove long-term trading skill or improvements yet.
What this means in practice
- •For trading platform operators: Identify discrepancies across decision, order, and execution steps in automated trading to improve audit transparency and compliance.
- •For algorithmic trading developers: Test and refine sequential trading agents by auditing internal consistency of decisions and execution traces with real market examples.
Authors
Arther Tian, Alex Ding, Simon Wu, Aaron Chan
Abstract
Sequential trading agents produce decisions, orders, execution receipts, account states, and reports that need not agree, even when the final portfolio value is correctly computed. We present FIDES-Live, an operational extension of FIDES from generated strategy concordance to the audit of sequential agent traces. The protocol separates structured decision validity, optional numerical commitment adherence, execution and ledger reconciliation, and receipt- grounded reporting, with explicit coverage and unscorable outcomes. We retrospectively audit ten requested inference routes over 7,299 cryptocurrency account-rounds and a 310-round equity pilot, using real market observations and local paper execution. Cryptocurrency decision validity is 97.35% and structured reporting accuracy is 90.81% among scorable responses, but route-level decision and reporting performance can diverge sharply. Commitment adherence is 94.36% overall; it falls to 62.63% on constraints emitted with nonempty order proposals, compared with 97.56% for valid empty proposals. All retained cash and position transitions reconcile with recorded fills, while execution controls and unavailable quotes explain separate rejection populations. Short-window marked returns provide descriptive context rather than a new test of whether concordance predicts alpha. These observations motivate reporting concordance, evidence coverage, activity, and financial outcomes separately. They do not establish long-run trading skill, semantic faithfulness of explanations, or causal benefits from audit feedback.