| | --- |
| | title: README |
| | emoji: 📚 |
| | colorFrom: red |
| | colorTo: indigo |
| | sdk: static |
| | pinned: true |
| | license: mit |
| | --- |
| | Welcome to the Manulife AI Research organization on Hugging Face! |
| |
|
| | At Manulife, we are committed to advancing the application of artificial intelligence to help people make better decisions and live better lives. Our AI research teams work across the organization — embedded within business and technology functions — to develop practical, responsible AI solutions that improve outcomes for our customers, advisors, and operations worldwide. |
| |
|
| | Our research spans key areas including **conversational AI and virtual assistants**, **retrieval-augmented generation for financial documents**, **LLM evaluation and benchmarking**, **causal inference and econometrics**, and **responsible AI governance**. We publish at leading academic venues and actively collaborate with university partners, including the University of Waterloo, to push the boundaries of AI in financial services. |
| |
|
| | We believe in building AI responsibly. Our [Responsible AI framework](https://manulife-ai.github.io/manulife-ai-research/) incorporates three independent lines of oversight, interpretable modeling practices, integrated bias testing, and privacy-preserving techniques — ensuring that our AI systems are trustworthy, fair, and transparent. |
| |
|
| | ### Open Source |
| |
|
| | We are proud to contribute to the open-source community with tools and datasets that advance AI research in finance: |
| |
|
| | - **[models-eval-playbook](https://github.com/manulife-ai/models-eval-playbook)** — A CLI-based framework for evaluating large language models across multiple providers (Azure OpenAI, Ollama, OpenRouter, and more) with MLflow integration for experiment tracking. |
| | - **[financialqa](https://github.com/manulife-ai/financialqa)** — A RAG-based question answering system for multi-structured financial documents, extracting and reasoning over text, tables, and figures from complex PDFs. |
| |
|
| | ### Selected Research |
| |
|
| | Our teams contribute to peer-reviewed research at the intersection of AI and financial services, with recent work in areas such as: |
| |
|
| | - Multi-structured document understanding using large language models (ECIR 2026) |
| | - Retrieval-augmented generation for financial institution knowledge bases |
| | - Fraud prevention and detection with generative AI |
| | - Analyst sentiment signal extraction and market dynamics modeling |
| | - Social media analysis for health and behavioral outcomes |
| |
|
| | ### Connect With Us |
| |
|
| | - 🌐 [Manulife AI Research](https://manulife-ai.github.io/manulife-ai-research/) |
| | - 💻 [GitHub](https://github.com/manulife-ai) |
| | - 📂 [Open Source Projects](https://manulife-ai.github.io/manulife-ai-research/open-source/) |
| |
|
| | [Learn more about AI Research at Manulife.](https://manulife-ai.github.io/manulife-ai-research/) |
