SpreadsheetBench v2
An agent evaluation benchmark focused on spreadsheet operations, requiring agents to understand table structure and execute formulas and data transformations, testing precise multi-step manipulation capabilities.
Timeline (last 90 days)
On the SpreadsheetBench v2 benchmark, EvoSkill v2 raises held-out pass rate by 7.7 percentage points over the stronger baseline.
SpreadsheetBench v2 focuses on spreadsheet manipulation, and τ³-bench banking belongs to the τ-bench family of tool-calling benchmarks simulating bank customer service scenarios.
All Facts (2)
On the SpreadsheetBench v2 benchmark, EvoSkill v2 raises held-out pass rate by 7.7 percentage points over the stronger baseline.
50%UnverifiedSpreadsheetBench v2 focuses on spreadsheet manipulation, and τ³-bench banking belongs to the τ-bench family of tool-calling benchmarks simulating bank customer service scenarios.
50%