SkillSafe-Bench

What is SkillSafe-Bench?

SkillSafe-Bench evaluates skill-merged LLMs on static refusal, adaptive jailbreak robustness, and capability retention using a two-judge AND rule, across…

Released

2026-08-09

Evaluates

General AI, Robustness, Safety, Safety & Trustworthiness

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.08542

SkillSafe-Bench paper

• Paper https://arxiv.org/abs/2608.08542

Which sources cite SkillSafe-Bench?

• SkillSafe-Bench

Related General AI, Robustness, Safety, Safety & Trustworthiness benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A3S-Bench
• AARRI-Bench
• ActBench
• ActiveSciBench
• ActiveVision