List your company

Spreadsheet Question Answering on FRTR-Bench full 1.0

74AccuracyFRTR
3.2821.644058.36Jan 13, 2026

Evaluation results

MethodLinks
FRTRBackbone=Claude Sonnet 4.52026.017411.71
FRTRBackbone=GPT-52026.017315.5
FRTRBackbone=Gemini 2.5 Pro2026.016726.9
FRTRBackbone=LLaMA Maverick-17B2026.01562.42
FRTRBackbone=GPT-OSS-120B2026.01515.75
FRTRBackbone=GPT-4o2026.01495.04
SpreadsheetLLMBackbone=Gemini 2.5 Pro2026.012433.35
SpreadsheetLLMBackbone=LLaMA Maverick-17B2026.01231.15
SpreadsheetLLMBackbone=GPT-OSS-120B2026.01218.8
SpreadsheetLLMBackbone=GPT-52026.011818.1
SpreadsheetLLMBackbone=Claude Sonnet 4.52026.01138.8
SpreadsheetLLMBackbone=GPT-4o2026.0161.14