Share your thoughts, 1 month free Claude Pro on usSee more

Contextual Integrity Applicability and Compliance on GoldCoin-HIPAA

96.3Applicability Macro F1

Phi-4

Updated 1mo ago

Evaluation Results

Method	Links
Phi-4 2026.04		96.3	67
ContextReasoner-7B 2026.04		96.3	57
CIRL-7B 2026.04		95.8	52.5
Qwen3.5-9B 2026.04		95.3	72.4
Gemma-3-12B 2026.04		94.4	74.4
Qwen3.5-9B 2026.04		93.9	75.3
Qwen3.5-9B 2026.04		93.4	73.3
Qwen3.5-4B 2026.04		90.1	73.1
Phi-4 2026.04		90.1	62
OpenThinker3-7B 2026.04		83	15.8
Gemma-3-12B 2026.04		80.1	74.4
Qwen3.5-2B 2026.04		79.3	44.8
GPT-OSS-20B 2026.04		78.6	73.4
OpenThinker3-7B 2026.04		77.6	15.8
Qwen3.5-4B 2026.04		69.6	72.3
GPT-OSS-20B 2026.04		63.9	58.2
Qwen3.5-2B 2026.04		56.2	58.6