EvoHarmBench

What is EvoHarmBench?

Evaluates content moderation systems against iteratively evolved adversarial samples across 229 semantic sub-clusters and five violation categories, with…

Released

2026-08-28

Evaluates

Factuality, General AI, Language & Knowledge, cs.CL

Openness

unknown

Importer

Claire Radar

Review status

ai-name-audit-deferred

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.27844

EvoHarmBench paper

• Paper https://arxiv.org/abs/2608.27844

Which sources cite EvoHarmBench?

• EvoHarmBench

Related Factuality, General AI, Language & Knowledge, cs.CL benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A Large-Scale Dataset and Benchmark
• A3S-Bench
• AARRI-Bench
• ABC-Bench
• ActBench