List your company

Instruction Following on MIF en

69.66AccuracyBaseline
60.799263.099665.467.7004Apr 29, 2026

Evaluation results

MethodLinks
BaselineFine-tuning=Prior to fine-tuning2026.0469.66
BaselineN=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.0469.63
DPON=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.0469.26
TLPOFine-tuning=After fine-tuning2026.0469.22
DPOFine-tuning=After fine-tuning2026.0467.31
ORPOFine-tuning=After fine-tuning2026.0466
TLPON=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.0465.76
ORPON=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.0465.72
SFTFine-tuning=After fine-tuning2026.0461.14
SFTN=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.0461.14