List your company

Spreadsheet Agent Skills on SpreadsheetBench (in-distribution)

48.33Vrf ScoreHuman-Written
-24.8132-5.824113.16532.1541Mar 26, 2026

Evaluation results

MethodLinks
Human-WrittenSkill User=Qwen3.5-122B-A10B2026.0348.3336.317.03
No SkillSkill User=Qwen3.5-122B-A10B2026.0327.6728.917.57
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-35B-A3B, Condition=Deepening (init: Human-Written), Refinement Component=+Error2026.03279.442.86
ParametricSkill User=Qwen3.5-122B-A10B2026.0326.1736.617.5
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Condition=Creation (init: Parametric), Refinement Component=+Error2026.0322.833.775.87
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Condition=Deepening (init: Human-Written), Refinement Component=+Combined2026.0321.510.8712.5
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-35B-A3B, Condition=Deepening (init: Human-Written), Refinement Component=+Combined2026.0321.168.841.8
ParametricSkill User=Qwen3.5-35B-A3B2026.0320.1713.73.87
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Condition=Deepening (init: Human-Written), Refinement Component=+Combined2026.03205.772.36
No SkillSkill User=Qwen3.5-35B-A3B2026.0319184.6
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Condition=Deepening (init: Human-Written), Refinement Component=+Error2026.0317.510.310.4
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Condition=Deepening (init: Human-Written), Refinement Component=+Error2026.0317.339.174.83
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-122B-A10B, Condition=Deepening (init: Human-Written), Refinement Component=+Error2026.0316.678.58.14
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Condition=Creation (init: Parametric), Refinement Component=+Success2026.0315.33-0.934.33
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-35B-A3B, Condition=Creation (init: Parametric), Refinement Component=+Success2026.0312.8311.576.13
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Condition=Deepening (init: Human-Written), Refinement Component=+Success2026.03113.640.83
Human-WrittenSkill User=Qwen3.5-35B-A3B2026.039.6713.033.37
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-35B-A3B, Condition=Deepening (init: Human-Written), Refinement Component=+Success2026.039.163.571.56
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-35B-A3B, Condition=Creation (init: Parametric), Refinement Component=+Error2026.038.669.534
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-122B-A10B, Condition=Deepening (init: Human-Written), Refinement Component=+Combined2026.036.673.874.17
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Condition=Creation (init: Parametric), Refinement Component=+Success2026.035.665.82.63
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-122B-A10B, Condition=Creation (init: Parametric), Refinement Component=+Success2026.035.33-4.572.43
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Condition=Creation (init: Parametric), Refinement Component=+Error2026.033.837.32.66
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-122B-A10B, Condition=Creation (init: Parametric), Refinement Component=+Error2026.031-7.71.03
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Condition=Creation (init: Parametric), Refinement Component=+Combined2026.030.16-9.23-1.4
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-35B-A3B, Condition=Creation (init: Parametric), Refinement Component=+Combined2026.03-0.174.41.26
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-122B-A10B, Condition=Creation (init: Parametric), Refinement Component=+Combined2026.03-0.84-9.17-1.63
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-35B-A3B, Condition=Creation (init: Parametric), Refinement Component=+Combined2026.03-1.173.731.36
Trace2SkillSkill Author=Qwen3.5-122B-A10B, Skill User=Qwen3.5-122B-A10B, Condition=Deepening (init: Human-Written), Refinement Component=+Success2026.03-21.83-8.570.04
Trace2SkillSkill Author=Qwen3.5-35B-A3B, Skill User=Qwen3.5-122B-A10B, Condition=Deepening (init: Human-Written), Refinement Component=+Success2026.03-22-8.83-0.5