List your company

Spreadsheet Reasoning on SpreadsheetBench Verified

71.11AccuracySkillGrad
31.818842.019452.2262.4206May 26, 2026

Evaluation results

MethodLinks
SkillGradBackbone=GPT-5.4, Initialization=LLM-gen., Training Type=Training-based skill improvement2026.0571.11
SkillGradBackbone=GPT-5.4, Initialization=Third-party, Training Type=Training-based skill improvement2026.0569.44
EvoSkillBackbone=GPT-5.4, Initialization=LLM-gen., Training Type=Training-based skill improvement2026.0568.06
Trace2SkillBackbone=GPT-5.4, Initialization=LLM-gen., Training Type=Training-based skill improvement2026.0565.28
Trace2SkillBackbone=GPT-5.4, Initialization=Third-party, Training Type=Training-based skill improvement2026.0563.89
EvoSkillBackbone=GPT-5.4, Initialization=Third-party, Training Type=Training-based skill improvement2026.0563.61
No SkillBackbone=GPT-5.4, Initialization=None, Training Type=Training-free2026.0562.5
Base xlsx SkillBackbone=GPT-5.4, Initialization=Third-party, Training Type=Training-free2026.0560
Base xlsx SkillBackbone=GPT-5.4, Initialization=LLM-gen., Training Type=Training-free2026.0555.83
SkillGradBackbone=GPT-4.1, Initialization=LLM-gen., Training Type=Training-based skill improvement2026.0554.17
SkillGradBackbone=GPT-4.1, Initialization=Third-party, Training Type=Training-based skill improvement2026.0545.83
No SkillBackbone=GPT-4.1, Initialization=None, Training Type=Training-free2026.0544.17
Trace2SkillBackbone=GPT-4.1, Initialization=Third-party, Training Type=Training-based skill improvement2026.0538.89
Trace2SkillBackbone=GPT-4.1, Initialization=LLM-gen., Training Type=Training-based skill improvement2026.0537.22
EvoSkillBackbone=GPT-4.1, Initialization=LLM-gen., Training Type=Training-based skill improvement2026.0537.22
EvoSkillBackbone=GPT-4.1, Initialization=Third-party, Training Type=Training-based skill improvement2026.0536.94
Base xlsx SkillBackbone=GPT-4.1, Initialization=LLM-gen., Training Type=Training-free2026.0536.67
Base xlsx SkillBackbone=GPT-4.1, Initialization=Third-party, Training Type=Training-free2026.0533.33