Unlocking Data Value in Finance: A Study on Distillation and Difficulty-Aware Training
This work provides high-quality datasets and models for researchers and practitioners working on data-centric financial AI, offering improved performance in a domain with low error tolerance.
This paper addresses the challenges of deploying Large Language Models (LLMs) in finance by demonstrating that performance is primarily driven by the quality and difficulty of post-training data. They introduce ODA-Fin-SFT-318k and ODA-Fin-RL-12k datasets, which enable an 8B model to consistently outperform open-source state-of-the-art financial LLMs on nine benchmarks.
Large Language Models (LLMs) have demonstrated strong general capabilities, yet their deployment in finance remains challenging due to dense domain-specific terminology, stringent numerical reasoning requirements, and low tolerance for factual errors. We conduct a controlled empirical study showing that in specialized vertical domains, performance is largely determined by the quality and difficulty/verifiability profile of post-training data. We introduce \textbf{ODA-Fin-SFT-318k}, constructed via multi-stage distillation and verification to produce high-quality Chain-of-Thought supervision, and \textbf{ODA-Fin-RL-12k}, curated for hard-but-verifiable tasks that balance reward precision and task diversity. Using standard SFT and RL pipelines, we show that high-quality CoT distillation establishes a robust foundation during SFT, while difficulty- and verifiability-aware sampling improves RL generalization. Evaluated on nine benchmarks spanning general financial tasks, sentiment analysis, and numerical reasoning, our ODA-Fin-RL-8B consistently surpasses open-source state-of-the-art (SOTA) financial LLMs of comparable size. We release our ODA-Fin-SFT-318k and ODA-Fin-RL-12k datasets, along with trained models to advance data-centric financial AI research.