Benchmark RadarAI BENCHMARK PROFILE
WADE
A reasoning-annotated benchmark for floating waste detection and grounding, with 2,167 images, 13,608 bounding boxes, and ten categories, each with recognition rules.
- Released
- 2026-08-24
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Provides a challenging benchmark for dense multi-instance grounding in a real-world environmental monitoring context, evaluating compact VLMs under various settings.
Motivation
Floating waste in inland waterways threatens aquatic ecosystems and requires timely monitoring under cluttered, multi-object conditions.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.