Spreadsheet Task Execution on SpreadsheetBench Verified (Held-out)
70.5Verification ScoreSkillCAT
Evaluation Results
| Method | Links | |
|---|---|---|
| SkillCATSkill Author=Qwen3.5-122B-A10B, Skill User=gemma-4-31B-it, Setting=Deepening2026.06 | 70.5 | |
| Trace2Skill +Comb.Skill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Setting=Deepening2026.06 | 69.83 | |
| SkillCATSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Setting=Deepening2026.06 | 69.5 | |
| SkillCATSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Setting=Creation2026.06 | 64.5 | |
| SkillCATSkill Author=Qwen3.5-35B-A3B, Skill User=gemma-4-31B-it, Setting=Deepening2026.06 | 61 | |
| SkillCATSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Setting=Deepening2026.06 | 55.5 | |
| SkillCATSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Setting=Creation2026.06 | 54.5 | |
| Human-WrittenSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Setting=Deepening2026.06 | 48.33 | |
| No-SkillSkill Author=Qwen3.5-35B-A3B, Skill User=gemma-4-31B-it, Setting=Deepening2026.06 | 46 | |
| No-SkillSkill Author=Qwen3.5-122B-A10B, Skill User=gemma-4-31B-it, Setting=Deepening2026.06 | 46 | |
| Human-WrittenSkill Author=Qwen3.5-35B-A3B, Skill User=gemma-4-31B-it, Setting=Deepening2026.06 | 39 | |
| Human-WrittenSkill Author=Qwen3.5-122B-A10B, Skill User=gemma-4-31B-it, Setting=Deepening2026.06 | 39 | |
| SkillCATSkill Author=Qwen3.5-35B-A3B, Skill User=gpt-5.4-mini, Setting=Deepening2026.06 | 37 | |
| SkillCATSkill Author=Qwen3.5-122B-A10B, Skill User=gpt-5.4-mini, Setting=Deepening2026.06 | 32.5 | |
| Human-WrittenSkill Author=Qwen3.5-35B-A3B, Skill User=gpt-5.4-mini, Setting=Deepening2026.06 | 31 | |
| Human-WrittenSkill Author=Qwen3.5-122B-A10B, Skill User=gpt-5.4-mini, Setting=Deepening2026.06 | 31 | |
| Trace2Skill +Comb.Skill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Setting=Deepening2026.06 | 29.67 | |
| Trace2Skill +Comb.Skill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Setting=Creation2026.06 | 26.33 | |
| ParametricSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Setting=Creation2026.06 | 26.17 | |
| ParametricSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Setting=Creation2026.06 | 20.17 | |
| Trace2Skill +Comb.Skill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Setting=Creation2026.06 | 20 | |
| Human-WrittenSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Setting=Deepening2026.06 | 9.67 | |
| No-SkillSkill Author=Qwen3.5-35B-A3B, Skill User=gpt-5.4-mini, Setting=Deepening2026.06 | 9 | |
| No-SkillSkill Author=Qwen3.5-122B-A10B, Skill User=gpt-5.4-mini, Setting=Deepening2026.06 | 9 |