CASTER-Bench: Human-Annotated Multimodal Benchmark for Community Resonance

CASTER-Bench is a human-annotated multimodal benchmark for Community-Aware Assessment of Social Textual Engagement and Resonance (CASTER). It was introduced by IndexTeam in a paper for ACL 2026 and is hosted on Hugging Face. The benchmark evaluates whether User-Generated Content achieves positive community resonance, moving beyond traditional aesthetic-focused Video Quality Assessment.

Use Cases

• Benchmarking models for community-aware assessment of social textual engagement based on the described task.
• Training models to evaluate positive resonance of User-Generated Content (UGC) beyond aesthetic quality.
• Researching the intersection of multimodal content and community feedback as described in the ACL 2026 paper.

Strengths

• Human-annotated benchmark, which suggests a level of quality control.
• Introduced in a peer-reviewed ACL 2026 paper, indicating academic rigor.
• Focuses on community resonance, a task described as going beyond traditional Video Quality Assessment.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset size are unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

IndexTeam via Hugging Face.

Collection Method

Human-annotated, as stated in the description.

Freshness

Last updated 2026-05-19 10:54:32; freshness should be verified.