AdversaBench

What is AdversaBench?

AdversaBench is an automated LLM red-teaming evaluation that mutates seed prompts and confirms resulting failures with multiple judges and a tiebreaker.

Released

2026-06-23

Evaluates

Cybersecurity, Language & Knowledge, cs.AI

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2606.24589

AdversaBench paper and code

• Paper https://arxiv.org/abs/2606.24589
• Code https://github.com/khanak0509/AdversaBench

Which sources cite AdversaBench?

• AdversaBench

Related Cybersecurity, Language & Knowledge, cs.AI benchmarks

• 5-Dialects-BN
• A Large-Scale Dataset and Benchmark
• AARRI-Bench
• ABC-Bench
• ActiveSciBench
• AdaMem-Bench
• AdvancedMathBench
• AdvPlan-Bench