LLMs as Judges: Resources for Automated GSN Assurance Case Review

Gerhard Yu created this repository of resources for research on LLMs as Judges: Toward The Automatic Review of GSN-compliant Assurance Cases. The dataset includes files in IPYNB, ZIP, CSV, and XLSX formats totaling 1.6 MB. It was last updated on May 5, 2026.

Use Cases

• Benchmarking LLM performance for automated review tasks based on the described research work.
• Developing tools for parsing and evaluating GSN-compliant assurance cases based on the repository's resources.
• Training or fine-tuning models for structured argument analysis based on the likely textual and tabular data.
• Reproducing or extending the research methodology described in 'LLMs as Judges'.

Strengths

• Includes multiple file formats (IPYNB, ZIP, CSV, XLSX) totaling 1.6 MB, suggesting varied resource types.
• Has a clear research context and purpose described in the title and description.
• Released under a CC-BY-4.0 license, facilitating open reuse and distribution.

Limitations

• Row count and column-level documentation are unknown, which limits suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Freshness

Last updated 2026-05-05 19:17:40; freshness should be verified.