Datasets:

---

IDEA-FinAI

/

Golden-Touchstone

like 2

Follow

IDEA FinAI 21

This PR updates the dataset card for the Golden Touchstone benchmark dataset. The updates aim to provide better context regarding its use and relation to the recent research presented in the paper Think-on-Graph 3.0: Efficient and Adaptive LLM Reasoning on Heterogeneous Graphs via Multi-Agent Dual-Evolving Context Retrieval and its associated RAG-Factory framework.

Key changes include:

• Adding task_categories: ['question-answering'] to improve discoverability.
• Adding language: ['en', 'zh'] to reflect its bilingual nature, as it's a "Comprehensive Bilingual Benchmark".
• Adding relevant tags such as rag, graph-neural-networks, llm-reasoning, financial, benchmark, and bilingual.
• Including a direct link to the Think-on-Graph 3.0 paper on Hugging Face.
• Adding a link to the RAG-Factory GitHub repository, which implements the concepts from the paper and can utilize this dataset.
• Incorporating the abstract of the Think-on-Graph 3.0 paper for better understanding.
• Adding details on RAG-Factory features, installation, and sample usage, extracted from the GitHub README.
• Ensuring the original citation for the Golden Touchstone dataset is retained, and adding a separate citation for the Think-on-Graph 3.0 paper for complete academic credit.