Benchmark RadarAI BENCHMARK PROFILE
CASTER-Bench
CASTER-Bench evaluates whether user-generated content achieves positive community resonance based on multimodal attributes, using human annotations across diverse categories.
- Released
- 2026-06-01
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Expands quality assessment from visual fidelity to social engagement, providing a task for modeling community resonance rather than just aesthetic quality.
Motivation
Traditional Video Quality Assessment (VQA) focuses narrowly on aesthetic fidelity, overlooking the complex social dynamics that define quality in User-Generated Content (UGC).
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.