Mathematics
MathConstraint
Tests whether models can determine if combinatorial constraints are satisfiable, with and without a solver.
504items
12subjects
cc-by-4.0license
mathematicsdomain
reasoningdomain
textmodality
item-level responses released
Saturation status: No
About this data
Adaptive combinatorial-reasoning benchmark: constraint-satisfaction (CSP/SAT) decision problems with solver-verified SAT/UNSAT ground truth; 12 models evaluated with and without a sandboxed SAT/SMT solver.
504 items · 11,862 recorded attempts. Browse item pages to see all observations. Filters and means apply to the current page.
Loading observation records…