Unlocking Data Value in Finance: A Study on Distillation and Difficulty-Aware Training
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 a...
| Authors | Chuxue Cao, Honglin Lin, Zhanping Zhong et al. |
|---|---|
| Primary category | cs.LG |
| Published | 7 March 2026 |
| Last updated at source | 7 March 2026 |
Connected in the knowledge graph
- OpenDataArena/ODA-Fin-RL-12kDataset · cited by
- OpenDataArena/ODA-Fin-SFT-318kDataset · cited by
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.