List your company

Reasoning on Shuffled Objects zero-shot

50.8AccuracyIn-Writing-Base
26.04832.47438.945.326Jan 12, 2026

Evaluation results

MethodLinks
In-Writing-BaseModel=Gemma2-9B-it, Prompting=Zero-shot, Variations=92026.0150.8
NL-to-FormatModel=Gemma2-9B-it, Prompting=Zero-shot, Variations=92026.0149.4
In-Writing-BaseModel=LLaMA3-8B-it, Prompting=Zero-shot, Variations=92026.0139
NL-to-FormatModel=LLaMA3-8B-it, Prompting=Zero-shot, Variations=92026.0127