Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2408.05365 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915024063168512 |
|---|---|
| author | Roychowdhury, Sohini Krema, Marko Moore, Brian Lai, Xingjian Effedua, Dike Jethwani, Bharat |
| author_facet | Roychowdhury, Sohini Krema, Marko Moore, Brian Lai, Xingjian Effedua, Dike Jethwani, Bharat |
| contents | Recent trends in Generative AI have emerged towards fine-tuning foundational large language models (LLMs) to create domain-specific LLMs for automation and chatbot-like applications. Specialized applications for analytics-heavy domains such as Financial report generation require specific writing styles that comprise compound and creative sentences with minimized hallucinations. In this work, we explore the self-corrective auto-regressive qualities of LLMs to learn creativity in writing styles with minimal prompting. We propose a novel two-stage fine-tuning (FT) strategy wherein in the first stage public domain financial reports are used to train for writing styles while allowing the LLM to hallucinate. In the second stage the examples of hallucinations are manually corrected and further used to fine-tune the LLM. The finally trained LLM learns to generate specific financial report sections using minimal instructions and tabular data inputs while ensuring low fine-tuning costs. Our proposed two-stage fine-tuning boosts the accuracy of financial questions answering by two-folds while reducing hallucinations by over 50%. Also, the fine-tuned model has lower perplexity, improved ROUGE, TER and BLEU scores, higher creativity and knowledge density with lower uncertainty and cross entropy than base LLMs. Thus, the proposed framework can be generalized to train creativity in LLMs by first allowing them to hallucinate. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_05365 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | FiSTECH: Financial Style Transfer to Enhance Creativity without Hallucinations in LLMs Roychowdhury, Sohini Krema, Marko Moore, Brian Lai, Xingjian Effedua, Dike Jethwani, Bharat Computation and Language Artificial Intelligence Computational Engineering, Finance, and Science Recent trends in Generative AI have emerged towards fine-tuning foundational large language models (LLMs) to create domain-specific LLMs for automation and chatbot-like applications. Specialized applications for analytics-heavy domains such as Financial report generation require specific writing styles that comprise compound and creative sentences with minimized hallucinations. In this work, we explore the self-corrective auto-regressive qualities of LLMs to learn creativity in writing styles with minimal prompting. We propose a novel two-stage fine-tuning (FT) strategy wherein in the first stage public domain financial reports are used to train for writing styles while allowing the LLM to hallucinate. In the second stage the examples of hallucinations are manually corrected and further used to fine-tune the LLM. The finally trained LLM learns to generate specific financial report sections using minimal instructions and tabular data inputs while ensuring low fine-tuning costs. Our proposed two-stage fine-tuning boosts the accuracy of financial questions answering by two-folds while reducing hallucinations by over 50%. Also, the fine-tuned model has lower perplexity, improved ROUGE, TER and BLEU scores, higher creativity and knowledge density with lower uncertainty and cross entropy than base LLMs. Thus, the proposed framework can be generalized to train creativity in LLMs by first allowing them to hallucinate. |
| title | FiSTECH: Financial Style Transfer to Enhance Creativity without Hallucinations in LLMs |
| topic | Computation and Language Artificial Intelligence Computational Engineering, Finance, and Science |
| url | https://arxiv.org/abs/2408.05365 |