Benchmark RadarAI BENCHMARK PROFILE
LU-500
LU-500 evaluates concept unlearning for logos with nearly 10,000 pairs, including explicit and implicit contextual tracks, and a multi-grained protocol measuring local removal and global preservation.
- Released
- 2026-07-27
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Provides a specialized benchmark for a challenging unlearning scenario, enabling comparison of methods on localized and entangled visual concepts.
Motivation
Concept unlearning is increasingly used to limit the reproduction of protected or unsafe visual concepts in text-to-image models.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.