Metraipolis › archive › dc_a67107f78f97
GRPO for Financial Advice Generation: Outperforming Commercial LLMs under CATE Evaluation — arXiv:2608.11787
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.11787
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T05:16:12.021Z
- SHA-256 of the archived text
7792a521da9458de1f057f0acc8a5d34b85d7e022a126948893f81a78619cecd- Stable URL of this page
https://metraipolis.com/documento/dc_a67107f78f97
How to cite
GRPO for Financial Advice Generation: Outperforming Commercial LLMs under CATE Evaluation — arXiv:2608.11787 — arXiv (en), https://arxiv.org/abs/2608.11787 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a67107f78f97 (Metraipolis archive dc_a67107f78f97, sha256 7792a521da9458de1f057f0acc8a5d34b85d7e022a126948893f81a78619cecd)
BibTeX
@misc{dca67107f78f97,
title = {GRPO for Financial Advice Generation: Outperforming Commercial LLMs under CATE Evaluation — arXiv:2608.11787},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.11787},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a67107f78f97; sha256 7792a521da9458de1f057f0acc8a5d34b85d7e022a126948893f81a78619cecd},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11787. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Generating actionable financial advice from business records demands that models integrate numerical reasoning, domain knowledge, and sound judgment, while avoiding recommendations that could harm the business. Direct supervision is difficult: historical decisions are not necessarily optimal, and high-quality free-form labels are expensive to obtain. We formulate financial advice generation as a reinforcement learning problem and fine-tune an open-weight language model using Group Relative Policy Optimization (GRPO). Our reward is an LLM-as-a-judge rubric that scores each recommendation across multiple binary dimensions of advice quality, augmented with a safety gate for harm prevention. Since LLM-based evaluation alone cannot confirm whether improvements reflect genuine business value rather than adaptation to the judge, we complement it with a judge-independent audit based on a standard doubly-robust Conditional Average Treatment Effect (CATE) estimator. Under this observational off-policy audit, our trained LLM achieves approximately twice the estimated gross-profit lift of the strongest evaluated commercial baseline ($0.0228$ vs.\ $0.0104$), together with the lowest downside rate and the least negative tail risk of any policy evaluated. Notably, the two evaluations do not rank the baselines identically: the untrained base model places last on the judge rubric but second on the causal audit, indicating that the audit captures a signal the judge does not. Our results demonstrate that GRPO with a finance-grounded reward signal can produce substantially more useful business recommendations than commercial LLMs, and that a judge-independent causal audit is a valuable complement to, rather than a confirmation of, LLM-as-a-judge assessment in financial NLP. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-13 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.11787 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-19T05:08:46.731Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).