IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2610.08882.html

FinVector-Market-4B: A Controlled Study of LoRA Adaptation for Structured Financial Tasks

Author

Listed:
  • Alina Khaybullina

Abstract

FinVector-Market-4B adapts Qwen/Qwen3.5-4B with rank-16 LoRA on a 22,000-example corpus for structured financial tasks. We evaluate the base and adapted models on the same 600-example benchmark under implicit and explicit JSON-schema contracts. Supplying the schema alone raises base-model JSON validity from 0% to 91.3%. Under matched explicit prompting, the frozen scores improve from 14.7% to 40.0% for FinQA answer exact match, from 48.0% to 82.7% for calculator-expression correctness, from 20.1% to 89.5% for scenario branch-label agreement, and from 52.4% to 87.2% for implication-direction agreement. A post-hoc policy-scoring audit shows that the reported macro-F1 decline reflects a changing label set; using the same three target classes gives 77.4% for the base and 83.1% for the adapter. Filing overlap and calculator-target inconsistencies qualify the benchmark's generalization claims. The results show that compact financial domain adaptation can produce substantial task-specific gains beyond output-format learning under matched prompting, with gains bounded by the evaluated task distribution and prompt contract.

Suggested Citation

  • Alina Khaybullina, 2026. "FinVector-Market-4B: A Controlled Study of LoRA Adaptation for Structured Financial Tasks," Papers 2610.08882, arXiv.org.
    • Handle: RePEc:arx:papers:2610.08882
    as

    Download full text from publisher

    File URL: https://arxiv.org/pdf/2610.08882
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Shijie Wu & Ozan Irsoy & Steven Lu & Vadim Dabravolski & Mark Dredze & Sebastian Gehrmann & Prabhanjan Kambadur & David Rosenberg & Gideon Mann, 2023. "BloombergGPT: A Large Language Model for Finance," Papers 2303.17564, arXiv.org, revised Dec 2023.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Mohammed-Khalil Ghali & Cecil Pang & Oscar Molina & Carlos Gershenson-Garcia & Daehan Won, 2025. "Forecasting Commodity Price Shocks Using Temporal and Semantic Fusion of Prices Signals and Agentic Generative AI Extracted Economic News," Papers 2508.06497, arXiv.org.
    2. Ching-Nam Hang & Pei-Duo Yu & Roberto Morabito & Chee-Wei Tan, 2024. "Large Language Models Meet Next-Generation Networking Technologies: A Review," Future Internet, MDPI, vol. 16(10), pages 1-29, October.
    3. Dangxing Chen & Pengzhan Guo, 2026. "Shapley in Context: Explaining Financial Language with Domain Expertise," Papers 2607.00856, arXiv.org.
    4. Xia Li & Hanghang Zheng & Xiwei Zhuang & Zhong Wang & Xiao Chen & Hong Liu & Jasmine Bai & Mao Mao, 2025. "Class-Imbalanced-Aware Adaptive Dataset Distillation for Scalable Pretrained Model on Credit Scoring," Papers 2501.10677, arXiv.org, revised Mar 2026.
    5. Lezhi Li & Ting-Yu Chang & Hai Wang, 2023. "Multimodal Gen-AI for Fundamental Investment Research," Papers 2401.06164, arXiv.org.
    6. Cristina Angelico & Enrico Bernardini, 2026. "Can GenAI fill banks' emissions data gaps?," Questioni di Economia e Finanza (Occasional Papers) 1003, Bank of Italy, Economic Research and International Relations Area.
    7. Amandeep Singh, 2026. "The Agentic ETF: How Agentic Trading Becomes an Asset Class," Papers 2610.06856, arXiv.org.
    8. Thanos Konstantinidis & Giorgos Iacovides & Mingxue Xu & Tony G. Constantinides & Danilo Mandic, 2024. "FinLlama: Financial Sentiment Classification for Algorithmic Trading Applications," Papers 2403.12285, arXiv.org.
    9. Yijia Xiao & Edward Sun & Tong Chen & Fang Wu & Di Luo & Wei Wang, 2025. "Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning," Papers 2509.11420, arXiv.org.
    10. Naysary, Babak & Edisen, Ali & Tarazi, Amine, 2026. "Climate risk news and banking industry: A natural language processing approach," Journal of Financial Stability, Elsevier, vol. 84(C).
    11. Meyer, Julian Anton, 2025. "Success factors and development areas for the implementation of Generative AI in companies," Junior Management Science (JUMS), Junior Management Science e. V., vol. 10(1), pages 1-23.
    12. Hui Gong, 2026. "AI Agents in Financial Markets: Architecture, Applications, and Systemic Implications," Papers 2603.13942, arXiv.org, revised Apr 2026.
    13. Turetken, Aysun Can & Leippold, Markus, 2026. "Battle of transformers: Adversarial attacks on financial sentiment models," Journal of Banking & Finance, Elsevier, vol. 188(C).
    14. Frank Xing, 2024. "Designing Heterogeneous LLM Agents for Financial Sentiment Analysis," Papers 2401.05799, arXiv.org.
    15. Maher Hamid, 2026. "Implementing domain-specific LLMs for strategic investment decisions: a retrospective case study comparing AI and human expertise," Digital Finance, Springer, vol. 8(1), pages 1-134, March.
    16. Yikuan Huang & Zheqi Fan & Kaiqi Hu & Yifan Ye, 2026. "From Hypotheses to Factors: Constrained LLM Agents in Cryptocurrency Markets," Papers 2604.26747, arXiv.org.
    17. Geofrey Ntale, 2026. "AI Trading: Evaluating Large Language Models for Technical Market Analysis," Papers 2607.15414, arXiv.org.
    18. Ankur Sinha & Chaitanya Agarwal & Pekka Malo, 2025. "FinBloom: Knowledge Grounding Large Language Model with Real-time Financial Data," Papers 2502.18471, arXiv.org, revised Feb 2026.
    19. Hoyoung Lee & Youngsoo Choi & Yuhee Kwon, 2024. "Quantifying Qualitative Insights: Leveraging LLMs to Market Predict," Papers 2411.08404, arXiv.org.
    20. Seppälä, Timo & Mucha, Tomasz & Mattila, Juri, 2023. "Beyond AI, Blockchain Systems, and Digital Platforms: Digitalization Unlocks Mass Hyper-Personalization and Mass Servitization," ETLA Working Papers 106, The Research Institute of the Finnish Economy.

    More about this item

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:arx:papers:2610.08882. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: arXiv administrators (email available below). General contact details of provider: https://arxiv.org/ .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.