Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Instruction Hierarchy Adherence on IHEval Aligned Single-turn
Loading...
84.8
Accuracy
Vanilla
40.3712
51.9056
63.44
74.9744
Jun 29, 2026
Accuracy
Updated 25d ago
Evaluation Results
Method
Method
Links
Accuracy
Vanilla
Model=Qwen3-8B-Instruct
2026.06
84.8
VerIH
Model=Qwen3-8B-Instruct
2026.06
84.59
Prompting
Model=Qwen3-8B-Instruct
2026.06
83.6
IHDec + VerIH
Model=Qwen3-8B-Instruct
2026.06
81.8
IHDec + Prompting
Model=Qwen3-8B-Instruct
2026.06
78.35
IHDec
Model=Qwen3-8B-Instruct
2026.06
77.2
IHDec
Model=Qwen3-4B-Instruct
2026.06
76.7
Prompting
Model=Qwen3-4B-Instruct
2026.06
75.9
IHDec + Prompting
Model=Qwen3-4B-Instruct
2026.06
75.9
IHDec
Model=Llama-3.1-8B-Ins...
2026.06
73.53
Vanilla
Model=Qwen3-4B-Instruct
2026.06
71
Vanilla
Model=Llama-3.1-8B-Ins...
2026.06
66.05
IHDec + Prompting
Model=Llama-3.1-8B-Ins...
2026.06
61.8
Prompting
Model=Llama-3.1-8B-Ins...
2026.06
56.5
IHDec + ISE
Model=Llama-3.1-8B-Ins...
2026.06
55.1
ISE
Model=Llama-3.1-8B-Ins...
2026.06
42.08
Feedback
Search any
task
Search any
task