Benchmark RadarAI BENCHMARK PROFILE
MiGUE-Bench
MiGUE-Bench is a benchmark for multi-granularity event analysis, covering event detection, relation reasoning, structure induction, and future prediction across single to cross-document settings. It uses an LLM-driven annotation pipeline.
- Released
- 2026-07-30
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Event analysis spans diverse tasks at different document granularity, and this benchmark aims to systematically assess LLM capabilities across them. Its breadth could inform progress in information extraction.
Motivation
Event analysis is an essential and fundamental direction of information extraction, involving various event-centric tasks at different granularity of documents.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.