Instruction-Following Evaluation for Large Language Models
Open paper
Benchmark
Model
Rank
Results
instruction-following-on-ifeval
GPT-4
#3
Inst-level loose-accuracy: 85.37
Inst-level strict-accuracy: 83.57
…
instruction-following-on-ifeval
PaLM 2 S
#4
Inst-level loose-accuracy: 59.11
Inst-level strict-accuracy: 55.76
…
Rank counts only results with a code link.