INS-MMBench: A Multimodal Benchmark for Insurance AI

INS-MMBench is the first comprehensive benchmark for evaluating Large Vision-Language Models in the insurance domain. It covers four insurance types—auto, property, health, and agricultural—and key insurance stages. The dataset was created by FDU-INS and was last updated on Hugging Face in July 2025.

Use Cases

• Benchmarking LVLM performance on insurance-specific tasks based on the described coverage of four insurance types.
• Training models to understand insurance documents and scenarios based on the multimodal question-answering format.
• Evaluating AI reasoning across key insurance stages mentioned in the dataset overview.

Strengths

• First benchmark specifically designed for the insurance domain for LVLMs.
• Covers four distinct insurance types: auto, property, health, and agricultural.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

FDU-INS

Collection Method

Proposed in an associated academic paper; specific collection method is not detailed in the provided input.

Freshness

Last updated 2025-07-10 11:43:24; freshness should be verified.