Dynamic Objective Selection with Safeguards and LLM Oversight for Financial Decision-Making
The paper introduces Dynamic Objective Selection with Safeguards (DOSS) for financial decision-making. DOSS aims to optimize trading objectives by adapting to changing market conditions without relying on intermediate regime variables. It incorporates a Large Language Model for oversight, ensuring that objective selections are both informed and conservative.
- ▪DOSS selects decision-relevant objective functions based on recent returns.
- ▪The system performs sequential updates to avoid temporal leakage in selections.
- ▪Confidence-aware gating is used to prevent misselection and excessive switching.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2606.03704 (cs) [Submitted on 2 Jun 2026] Title:Dynamic Objective Selection with Safeguards and LLM Oversight for Financial Decision-Making Authors:Keigo Sakurai, Takahiro Ogawa, Miki Haseyama, Anjyu Anan, Kei Nakagawa View a PDF of the paper titled Dynamic Objective Selection with Safeguards and LLM Oversight for Financial Decision-Making, by Keigo Sakurai and 4 other authors View PDF HTML (experimental) Abstract:Financial decision-making tasks such as stock recommendation and portfolio allocation typically estimate future return and risk and then select trades or allocations for an investor, and the chosen optimization objective often determines realized performance.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
