Newest papers

AllPath: Layered Verification for a Deployed, Self-Revising LLM Trading Agent under Human-Gated and Autonomous Regimes

Tian Zhang, Meng Li · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26662 · cited by 0 works (OpenAlex) · open-access PDF

Korean

LLM 컨트롤러는 자신의 조건을 통한 수동 거래 에이전트를 거의 독점적으로 백 테스트 및 자율적 인 벤치 마크에 평가합니다. 조치가 돌이킬 수없는 배포는 다른 질문을 제기합니다 : 누가 에이전트의 행동을 검증하고 어떤 실패를 각 검사자가 구조적으로 잡을 수 있습니까? 우리는 AllPath에 보고합니다. 개인 LLM 거래 에이전트는 라이브 종이 브로커링 계정에 위험 제한을 적용하고 통제하지 않는 모든 전략을 차단합니다. 시스템은 결정적인 규칙 엔진을 통해 위치를 모니터링합니다. 에이전트는 정해진 규칙 내에서 자신을 검사합니다. 매일 밤 그것은 자신의 거래 전략에 대한 검토를 제안합니다. 검증은 세

Machine translation (Meta M2M100); the English original follows.

Abstract

LLM trading agents are evaluated almost exclusively in backtests and autonomous benchmark harnesses. Deployment, where actions are irreversible, raises a different question: who verifies the agent's actions, and which failures can each verifier structurally catch? We report on AllPath, a personal LLM trading agent deployed against a live paper-brokerage account. The system watches positions through a deterministic rules engine, and each night it reviews its own day and proposes revisions to its own trading strategies. Verification is split across three layers. Code verifies what must not depend on judgment: deterministic gates enforce risk limits and block any strategy revision that fails an integrity check. The agent verifies itself within fixed bounds: its nightly review has an iteration cap and no tools that can act. A human verifies judgment, through approval screens that show the current state of each proposal. We also compared two verification policies for the agent's self-revisions on the live system: two and a half weeks with every self-revision human-gated, then five weeks with reflection revisions auto-applied behind the same code gates, while the operator's chat channel stayed human-gated throughout. Each layer caught a different kind of failure. The code gates rejected no proposal; their role was keeping revisions consistent. The human rejected two proposals that passed every code check, both over a disagreement about the market, and later removed 11 of the 19 rules the agent had added on its own, at a median lag of eight days. The most instructive production incidents, such as strategies silently left unmonitored, passed every gate; the operator found them by hand, and the fix each time was an interface change that makes the condition visible. We contribute the layered architecture, the two-regime account, and a production incident taxonomy, none observable in a backtest.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Code (set theory), Computer science, Benchmark (surveying), State (computer science), Channel (broadcasting), Reflection (computer programming)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange