Empowering Small Language Models with Factual Hallucination-Aware Reasoning for Financial Classification
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914232993316864 |
|---|---|
| author | Yuan, Han Wu, Yilin Zhang, Li Ma, Zheng |
| author_facet | Yuan, Han Wu, Yilin Zhang, Li Ma, Zheng |
| contents | Small language models (SLMs) are increasingly used for financial classification due to their fast inference and local deployability. However, compared with large language models, SLMs are more prone to factual hallucinations in reasoning and exhibit weaker classification performance. This raises a natural question: Can mitigating factual hallucinations improve SLMs' financial classification? To address this, we propose a three-step pipeline named AAAI (Association Identification, Automated Detection, and Adaptive Inference). Experiments on three representative SLMs reveal that: (1) factual hallucinations are positively correlated with misclassifications; (2) encoder-based verifiers effectively detect factual hallucinations; and (3) incorporating feedback on factual errors enables SLMs' adaptive inference that enhances classification performance. We hope this pipeline contributes to trustworthy and effective applications of SLMs in finance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_01378 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Empowering Small Language Models with Factual Hallucination-Aware Reasoning for Financial Classification Yuan, Han Wu, Yilin Zhang, Li Ma, Zheng Artificial Intelligence Small language models (SLMs) are increasingly used for financial classification due to their fast inference and local deployability. However, compared with large language models, SLMs are more prone to factual hallucinations in reasoning and exhibit weaker classification performance. This raises a natural question: Can mitigating factual hallucinations improve SLMs' financial classification? To address this, we propose a three-step pipeline named AAAI (Association Identification, Automated Detection, and Adaptive Inference). Experiments on three representative SLMs reveal that: (1) factual hallucinations are positively correlated with misclassifications; (2) encoder-based verifiers effectively detect factual hallucinations; and (3) incorporating feedback on factual errors enables SLMs' adaptive inference that enhances classification performance. We hope this pipeline contributes to trustworthy and effective applications of SLMs in finance. |
| title | Empowering Small Language Models with Factual Hallucination-Aware Reasoning for Financial Classification |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2601.01378 |