Search arXivSearch

arXiv · 2511.08621

The LLM Pro Finance Suite: Multilingual Large Language Models for Financial Applications

Abstract

The financial industry's growing demand for advanced natural language processing (NLP) capabilities has highlighted the limitations of generalist large language models (LLMs) in handling domain-specific financial tasks. To address this gap, we introduce the LLM Pro Finance Suite, a collection of five instruction-tuned LLMs (ranging from 8B to 70B parameters) specifically designed for financial applications. Our approach focuses on enhancing generalist instruction-tuned models, leveraging their existing strengths in instruction following, reasoning, and toxicity control, while fine-tuning them on a curated, high-quality financial corpus comprising over 50% finance-related data in English, French, and German. We evaluate the LLM Pro Finance Suite on a comprehensive financial benchmark suite, demonstrating consistent improvement over state-of-the-art baselines in finance-oriented tasks and financial translation. Notably, our models maintain the strong general-domain capabilities of their base models, ensuring reliable performance across non-specialized tasks. This dual proficiency, enhanced financial expertise without compromise on general abilities, makes the LLM Pro Finance Suite an ideal drop-in replacement for existing LLMs in financial workflows, offering improved domain-specific performance while preserving overall versatility. We publicly release two 8B-parameters models to foster future research and development in financial NLP applications: https://huggingface.co/collections/DragonLLM/llm-open-finance.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Gaëtan Caillaut, Raheel Qader, Jingshu Liu, Mariam Nakhlé, Arezki Sadoune, Massinissa Ahmim, Jean-Gabriel Barthelemy. 2025-11-07. The LLM Pro Finance Suite: Multilingual Large Language Models for Financial Applications. https://arxiv.org/abs/2511.08621

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

The Endogenous Constraint: Hysteresis, Stagflation, and the Structural Inhibition of Monetary Velocity in the Bitcoin Network (2016-2025)

Bitcoin operates as a macroeconomic paradox: it combines a strictly predetermined, inelastic monetary issuance schedule with a stochastic, highly elastic demand for scarce block space. This paper empirically validates the Endogenous Constraint Hypothesis, positing that protocol-level throughput limits generate a non-linear negative feedback loop between network friction and base-layer monetary velocity. Using a verified Transaction Cost Index (TCI) derived from Blockchain.com on-chain data and Hansen's (2000) threshold regression, we identify a definitive structural break at the 90th percentile of friction (TCI ~ 1.63). The analysis reveals a bifurcation in network utility: while the network exhibits robust velocity growth of +15.44% during normal regimes, this collapses to +6.06% during shock regimes, yielding a statistically significant Net Utility Contraction of -9.39% (p = 0.012). Crucially, Instrumental Variable (IV) tests utilizing Hashrate Variation as a supply-side instrument fail to detect a significant relationship in a linear specification (p=0.196), confirming that the velocity constraint is strictly a regime-switching phenomenon rather than a continuous linear function. Furthermore, we document a "Crypto Multiplier" inversion: high friction correlates with a +8.03% increase in capital concentration per entity, suggesting that congestion forces a substitution from active velocity to speculative hoarding.

q-fin.ST

Wasserstein-Barycentric Interaction Fields for Spatial Factor Models: Evidence from Language-Model Representations

Spatial asset-pricing models take the structure of inter-firm interaction as given. We infer that structure from firms' information environments using language-model representations. Each firm is represented as a distribution of news-article embeddings, and a target-anchored Wasserstein barycentric reconstruction selects, for every firm, the weighted combination of other firms whose information footprints jointly reconstruct its own. The resulting directed peer field enters a quadratic exposure-adjustment model in which the spatial coefficient indexes alignment with information peers relative to stand-alone exposure. Using fields built from 2018-2022 news and frozen before 2023-2026 returns, we find that the constructed field organizes cross-sectional return dependence beyond the Fama-French five factors and momentum and raises the held-out mean Gaussian quasi-log score relative to a matched factor-only model. Because factor betas are unchanged, the gain lies in residual covariance. The field outperforms pairwise distance weighting and equal weighting of the same peers, and remains incrementally informative beside persistent news co-mentions under the primary factor-conditioned specification. Linear and quadratic transport generate nearly identical peer-return signals and equivalent held-out predictive performance. The barycentric-proximity ordering persists across alternative embedding models, and a pre-period encoder preserves the held-out advantage under the primary specification. Language-model representations thus serve as a measurement instrument for latent inter-firm information structure in capital markets.

q-fin.ST

Stealing profits: Spread-based temporal hierarchy forecasting for day-ahead electricity markets

Day-ahead electricity price forecasts support trading and storage decisions, but for battery arbitrage predicting intraday price spreads is more relevant than predicting individual hourly prices. Here we show that a temporal hierarchy forecasting (THieF) framework that jointly reconciles forecasts of hourly electricity prices and all intraday price spreads consistently improves performance across two major European electricity markets and three different forecasting architectures. Using five years of out-of-sample data from Germany and Spain, we obtain accuracy improvements of up to 19.7% and profit gains of up to 10.4% relative to unreconciled hourly price forecasts. The gains persist even for a highly accurate pretrained TabPFN foundation model. Our results demonstrate that exploiting coherent relationships between economically relevant forecasting targets can improve both predictive accuracy and decision value, and that better statistical forecasts do not necessarily imply better economic decisions.

q-fin.ST