SafePyramid Benchmark

What is SafePyramid?

SafePyramid evaluates in-context policy guardrailing across 1,000 multi-turn conversations and 3,000 application-specific policies containing 61,699…

Released

2026-06-29

Evaluates

General AI, Safety, Safety & Trustworthiness

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2606.29887

SafePyramid paper and code

• Paper https://arxiv.org/abs/2606.29887
• Code https://github.com/bytedance/safepyramid

Which sources cite SafePyramid?

• SafePyramid

Related General AI, 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