## 🚀 最新动态
- **2026.05.20** 🎉 [**DianJin-SKILLS (百技图)**](DianJin-SKILLS/README.md) 正式开源 —— 面向金融领域的 AI Agent 技能库，涵盖银行、保险和证券/资管方向：**10 个专业角色，130+ 标准化技能**，可随时接入各类 Agent 框架。
- **2026.05.10** 🎉 "[Fin-PRM: A Domain-Specialized Process Reward Model for Financial Reasoning in Large Language Models](https://arxiv.org/abs/2508.15202)" 被 **IJCAI 2026** 收录！
- **2026.04.07** 🎉 "[Benchmarking Large Vision-Language Models on CFMME: A Comprehensive Chinese Financial Multimodal Evaluation Dataset](https://arxiv.org/abs/2605.29462)" 被 **ACL 2026** 主会收录！
- **2026.02.27** 🎉 我们的论文 **FinMCP-Bench** 和 **CARE** 被 **ICASSP 2026** 收录！
- **2025.11.15** "[Evaluating, Synthesizing, and Enhancing for Customer Support Conversation](https://arxiv.org/abs/2508.04423)" 正式被 **AAAI 2026** 收录！
- **2025.10.11** "[FinMCP-Bench: Benchmarking LLM Agents for Real-World Financial Tool Use under the Model Context Protocol](./DianJin-TIR/technical%20report_FinMCP_Bench.pdf)" 由盈米基金及合作伙伴联合发布 —— 这是首个基于 MCP 构建、用于评估 LLM agent 在真实金融场景中工具调用能力的基准数据集与评测框架。
- **2025.10.11** "[CARE: Cognitive-reasoning Augmented Reinforcement for Emotional Support Conversation](https://arxiv.org/abs/2510.05122)" 现已发布！
- **2025.08.25** "[Fin-PRM: A Domain-Specialized Process Reward Model for Financial Reasoning in Large Language Models](https://arxiv.org/abs/2508.15202)" 论文现已发布并开源！
- **2025.08.18** "[DianJin-OCR-R1: Enhancing OCR Capabilities via a Reasoning-and-Tool Interleaved Vision-Language Model](https://www.arxiv.org/abs/2508.13238)" 论文现已发布并开源！
- **2025.08.08** "[Evaluating, Synthesizing, and Enhancing for Customer Support Conversation](https://arxiv.org/abs/2508.04423)" 论文现已发布并开源！
更早动态
- **2025.05.22** "[M3FinMeeting: A Multilingual, Multi-Sector, and Multi-Task Financial Meeting Understanding Evaluation Dataset](https://arxiv.org/abs/2506.02510)" 正式被 ACL 2025 收录！
- **2025.04.23** [DianJin-R1](DianJin-R1/README.md) 系列正式开源！包含 DianJin-R1-Data 数据集以及两款强大模型 DianJin-R1-7B 和 DianJin-R1-13B。详情请参阅我们的技术报告 "[DianJin-R1: Evaluating and Enhancing Financial Reasoning in Large Language Models](https://arxiv.org/abs/2504.15716)"。
- **2025.01.06** [CFLUE](https://github.com/aliyun/cflue) 数据集已全面开源，现已开放下载！🚀🚀🚀
- **2024.05.16** "[Benchmarking Large Language Models on CFLUE - A Chinese Financial Language Understanding Evaluation Dataset](https://arxiv.org/abs/2405.10542)" 正式被 ACL 2024 收录！🚀🚀🚀
## 📦 开源项目
本仓库是我们金融 AI 研究的开源枢纽。已发布的**代码**、**模型**和**数据**汇总如下：
## 📝 简介
欢迎来到 **Qwen DianJin** 👋
我们是来自 **阿里云金融行业** 的 **通义金融大模型团队**，专注于大模型在金融领域的探索、研究与部署。
### 方法论：DianJin 数据飞轮
我们的工作遵循 **评估 → 数据合成 → Post-Training** 的闭环方法论，每一次循环都会产出可复用的金融能力，并反哺至下一轮：
- **评估** 通过行业级 benchmark 设定可量化的能力目标。
- **数据合成** 通过可控、可扩展的轨迹数据，将金融领域长期存在的“数据稀缺”难题转化为**“数据生产力”**。
- **Post-Training** (SFT, RL) 将数据转化为稳定、可迁移的任务能力，进而再次反哺评估与数据生成环节。
### 研究方向
- **金融 benchmark 与评估** —— 构建涵盖基础能力 (CFLUE)、场景能力 (M
3FinMeeting) 和 Agent 能力 (FinMCP-Bench) 的三层能力框架，并共建 FinGDPVal。
- **数据合成** —— 基于 Agent 和 workflow 的轨迹蒸馏、MCP 环境交互以及多智能体自博弈；为 DianJin-R1-Data 和 CSC 语料库等数据集提供动力。
- **Post-Training** —— DianJin-R1 推理模型，以及首个金融领域 process reward model Fin-PRM。
- **智能对话** —— 具备显式认知-情感-策略推理链的客服支持 (CSC) 和情感支持 (CARE) 对话。
- **Agentic** —— DianJin-SKILLS (百技图) 技能库与 DianJin-TIR 自主规划与工具调用。
- **多模态** —— 用于金融文档理解的推理与工具交织 VLMs (DianJin-OCR-R1)。
**2026 年重点：** Agentic RL 与 meta-harness 演进。
本仓库即是上述努力的开源大本营。
## ✨ 托管平台
我们的研究成果已在 [**Qwen DianJin 平台**](https://tongyi.aliyun.com/dianjin) 落地产品化 —— 这是阿里云面向金融机构的 AI 工作台。欢迎访问平台官网了解产品详情并获取使用权限。
我们还在与行业合作伙伴共建 **FinGDPVal** —— 一个旨在评估 AI 工作台能否端到端完成真实金融工作任务的评测 benchmark。其黄金标准评测集与工具链将在此开源。
## 🔖 引用
如果我们的工作对您有帮助，欢迎引用我们。
```
@inproceedings{csconv,
title = {Evaluating, Synthesizing, and Enhancing for Customer Support Conversation},
author = {Jie Zhu and Huaixia Dou and Junhui Li and Lifan Guo and Feng Chen and Chi Zhang and Fang Kong},
booktitle = {Proceedings of AAAI},
year = {2026}
}
@inproceedings{fin-prm,
title = {Fin-PRM: A Domain-Specialized Process Reward Model for Financial Reasoning in Large Language Models},
author = {Jie Zhu and Yuanchen Zhou and Shuo Jiang and Junhui Li and Lifan Guo and Feng Chen and Chi Zhang},
booktitle = {Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI)},
year = {2026}
}
@inproceedings{finmcp-bench,
title = {FinMCP-Bench: Benchmarking LLM Agents for Real-World Financial Tool Use under the Model Context Protocol},
author = {Jie Zhu and Yimin Tian and Boyang Li and Kehao Wu and Zhongzhi Liang and Junhui Li and Xianyin Zhang and Lifan Guo and Feng Chen and Yong Liu and Chi Zhang},
booktitle = {Proceedings of ICASSP},
year = {2026},
pages = {19782--19786}
}
@inproceedings{care-esc,
title = {CARE: Cognitive-Reasoning Augmented Reinforcement for Emotional Support Conversation},
author = {Jie Zhu and Yuanchen Zhou and Shuo Jiang and Junhui Li and Lifan Guo and Feng Chen and Chi Zhang and Fang Kong},
booktitle = {Proceedings of ICASSP},
year = {2026},
pages = {17547--17551}
}
@article{dianjin-ocr-r1,
title = {DianJin-OCR-R1: Enhancing OCR Capabilities via a Reasoning-and-Tool Interleaved Vision-Language Model},
author = {Qian Chen and Xianyin Zhang and Lifan Guo and Feng Chen and Chi Zhang},
journal = {arXiv preprint arXiv:2508.13238},
year = {2025}
}
@inproceedings{m3finmeeting,
title = {M$^3$FinMeeting: A Multilingual, Multi-Sector, and Multi-Task Financial Meeting Understanding Evaluation Dataset},
author = {Jie Zhu and Junhui Li and Yalong Wen and Xiandong Li and Lifan Guo and Feng Chen},
booktitle = {Findings of the Association for Computational Linguistics: ACL 2025},
year = {2025},
pages = {244--266}
}
@article{dianjin-r1,
title = {DianJin-R1: Evaluating and Enhancing Financial Reasoning in Large Language Models},
author = {Jie Zhu and Qian Chen and Huaixia Dou and Junhui Li and Lifan Guo and Feng Chen and Chi Zhang},
journal = {arXiv preprint arXiv:2504.15716},
year = {2025}
}
@inproceedings{cflue,
title = {Benchmarking Large Language Models on CFLUE - A Chinese Financial Language Understanding Evaluation Dataset},
author = {Jie Zhu and Junhui Li and Yalong Wen and Lifan Guo},
booktitle = {Findings of the Association for Computational Linguistics: ACL 2024},
year = {2024},
pages = {5673--5693}
}
```
## 🤝 联系我们
感谢您对通义金融大模型系列的兴趣！如有研究或产品方面的咨询，请通过邮件 **CFLUE@alibabacloud.com** 与我们的团队取得联系，或扫描下方二维码加入我们的钉钉群。
## ⚠️ 免责声明
我们对使用 DianJin 开源模型和数据的行为不承担任何法律责任。用户需自行评估并承担任何潜在风险，验证模型输出，并针对自身的应用场景做出审慎决策。发布的数据和模型仅用于学术研究与行业应用，旨在推动 AI 在数据分析、金融创新及相关领域的发展。
