QianfanHuijin Technical Report: A Novel Multi-Stage Training Paradigm for Finance Industrial LLMs
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918294564372480 |
|---|---|
| author | Li, Shupeng Lu, Weipeng Liu, Linyun Lin, Chen Li, Shaofei Tan, Zhendong Zhong, Hanjun Zeng, Yucheng Zhu, Chenghao Liu, Mengyue Dong, Daxiang Wu, Jianmin Xiao, Yunting Li, Annan Liu, Danyu Zhang, Jingnan Liu, Licen Yin, Dawei Shen, Dou |
| author_facet | Li, Shupeng Lu, Weipeng Liu, Linyun Lin, Chen Li, Shaofei Tan, Zhendong Zhong, Hanjun Zeng, Yucheng Zhu, Chenghao Liu, Mengyue Dong, Daxiang Wu, Jianmin Xiao, Yunting Li, Annan Liu, Danyu Zhang, Jingnan Liu, Licen Yin, Dawei Shen, Dou |
| contents | Domain-specific enhancement of Large Language Models (LLMs) within the financial context has long been a focal point of industrial application. While previous models such as BloombergGPT and Baichuan-Finance primarily focused on knowledge enhancement, the deepening complexity of financial services has driven a growing demand for models that possess not only domain knowledge but also robust financial reasoning and agentic capabilities. In this paper, we present QianfanHuijin, a financial domain LLM, and propose a generalizable multi-stage training paradigm for industrial model enhancement.
Our approach begins with Continual Pre-training (CPT) on financial corpora to consolidate the knowledge base. This is followed by a fine-grained Post-training pipeline designed with increasing specificity: starting with Financial SFT, progressing to Finance Reasoning RL and Finance Agentic RL, and culminating in General RL aligned with real-world business scenarios. Empirical results demonstrate that QianfanHuijin achieves superior performance across various authoritative financial benchmarks. Furthermore, ablation studies confirm that the targeted Reasoning RL and Agentic RL stages yield significant gains in their respective capabilities. These findings validate our motivation and suggest that this fine-grained, progressive post-training methodology is poised to become a mainstream paradigm for various industrial-enhanced LLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_24314 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | QianfanHuijin Technical Report: A Novel Multi-Stage Training Paradigm for Finance Industrial LLMs Li, Shupeng Lu, Weipeng Liu, Linyun Lin, Chen Li, Shaofei Tan, Zhendong Zhong, Hanjun Zeng, Yucheng Zhu, Chenghao Liu, Mengyue Dong, Daxiang Wu, Jianmin Xiao, Yunting Li, Annan Liu, Danyu Zhang, Jingnan Liu, Licen Yin, Dawei Shen, Dou Computation and Language Domain-specific enhancement of Large Language Models (LLMs) within the financial context has long been a focal point of industrial application. While previous models such as BloombergGPT and Baichuan-Finance primarily focused on knowledge enhancement, the deepening complexity of financial services has driven a growing demand for models that possess not only domain knowledge but also robust financial reasoning and agentic capabilities. In this paper, we present QianfanHuijin, a financial domain LLM, and propose a generalizable multi-stage training paradigm for industrial model enhancement. Our approach begins with Continual Pre-training (CPT) on financial corpora to consolidate the knowledge base. This is followed by a fine-grained Post-training pipeline designed with increasing specificity: starting with Financial SFT, progressing to Finance Reasoning RL and Finance Agentic RL, and culminating in General RL aligned with real-world business scenarios. Empirical results demonstrate that QianfanHuijin achieves superior performance across various authoritative financial benchmarks. Furthermore, ablation studies confirm that the targeted Reasoning RL and Agentic RL stages yield significant gains in their respective capabilities. These findings validate our motivation and suggest that this fine-grained, progressive post-training methodology is poised to become a mainstream paradigm for various industrial-enhanced LLMs. |
| title | QianfanHuijin Technical Report: A Novel Multi-Stage Training Paradigm for Finance Industrial LLMs |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2512.24314 |