Benchmark RadarAI BENCHMARK PROFILE
MolSafeEval
MolSafeEval evaluates safety risks in AI-generated molecules across four task types, using structured knowledge graphs and LLM-based reasoning to detect unsafe features.
- Released
- 2026-07-01
- Readiness
- Paper only
- Primary field
- Science & Research
Why it matters
Fills a gap in molecular generation benchmarks by focusing on safety risks, providing standardized datasets and protocols for safer molecular design.
Motivation
Current molecular generation benchmarks emphasize task complexity, molecule novelty, and property alignment; they largely overlook a critical concern: the potential safety risks of AI-generated molecules.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.