aiqbench analysis
AlphaDiverse addresses critical limitations in LLM-based financial analysis by shifting from external API dependencies to local, post-trained agents that maintain cost control and data confidentiality. By employing a joint GRPO optimization method on diverse research paths, the framework prevents model collapse into repetitive economic mechanisms while enhancing exploratory breadth. This approach demonstrates how multi-agent systems can effectively balance competitive predictive accuracy with robust diversification in quantitative factor mining.
AlphaDiverse: Post-Training Local Quantitative Research Agents for Diverse Exploration in Alpha Factor Mining
arXiv:2609.29014v1 Announce Type: new Abstract: Large language model (LLM)-based multi-agent systems can automate alpha factor mining, but their reliance on external APIs limits control over cost, availability, and confidentiality. Long research loops also tend to revisit a few successful economic mechanisms that lead to research path collapse. To address these limitations, we propose AlphaDiverse, a framework that integrates a multi-agent alpha research system, diverse research path collection, and post-training for local agents. We let the research system generate complementary plan portfolios and vary research environments across loops to collect diverse research paths. Using these diverse traces, we warm-start local Planner and Realizer agents with supervised fine-tuning. Then, we propose a joint GRPO method to optimize both of them using predictive quality and diversity of contributions. Research feedback is confined to inner period data, while a frozen final model is evaluated on a later outer period data, thereby avoiding test-set tuning. Experiments across four Chinese stock universes show that AlphaDiverse can combine competitive prediction with broader exploration.
Abstract:Large language model (LLM)-based multi-agent systems can automate alpha factor mining, but their reliance on external APIs limits control over cost, availability, and confidentiality. Long research loops also tend to revisit a few successful economic mechanisms that lead to research path collapse. To address these limitations, we propose AlphaDiverse, a framework that integrates a multi-agent alpha research system, diverse research path collection, and post-training for local agents. We let the research system generate complementary plan portfolios and vary research environments across loops to collect diverse research paths. Using these diverse traces, we warm-start local Planner and Realizer agents with supervised fine-tuning. Then, we propose a joint GRPO method to optimize both of them using predictive quality and diversity of contributions. Research feedback is confined to inner period data, while a frozen final model is evaluated on a later outer period data, thereby avoiding test-set tuning. Experiments across four Chinese stock universes show that AlphaDiverse can combine competitive prediction with broader exploration.
33 pages, 7 figures, 26 Tables. Preprint under review
Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Multiagent Systems (cs.MA)
arXiv:2609.29014 [cs.AI]
(or arXiv:2609.29014v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.29014 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 273 words
Related coverage
[AINews] Opus 5.5 is good at explainer videos
Latent Space · Sep 29, 2026
[AINews] The Future of Latent Space
Latent Space · Sep 25, 2026
Expenditure Horizon: Measuring Optimization Ability, with an Application to NanoGPT
METR · Jul 21, 2026
Game-Theoretic Disempowerment: Loss of Control to Agencyless AI
LessWrong · Sep 27, 2026