ctxdexSynthetic model data for learning

Benchmark family

MathArena BrokenArXiv

A family groups the versions of one benchmark. It is not evaluated and has no applicable models of its own; each of its 6 definitions is resolved separately.

General reasoningRolling evaluationResearch mathematics

Definitions

Compare the newest 4 →

Newest first, following the recorded successor chain. Each definition is resolved against models separately.

  1. BenchmarkMathArena BrokenArXiv08/2026
    Active
    Successor of MathArena BrokenArXiv
    Scale
    56 problems
    Primary metric
    Mean judged score
    Applicable models
    13
  2. BenchmarkMathArena BrokenArXiv06/2026
    Active
    Successor of MathArena BrokenArXiv
    Scale
    54 problems
    Primary metric
    Mean judged score
    Applicable models
    13
  3. BenchmarkMathArena BrokenArXiv05/2026
    ActiveNewest in family
    Successor of MathArena BrokenArXiv
    Scale
    50 problems
    Primary metric
    Mean judged score
    Applicable models
    13
  4. BenchmarkMathArena BrokenArXiv04/2026
    Deprecated
    Successor of MathArena BrokenArXiv
    Scale
    61 problems
    Primary metric
    Mean judged score
    Applicable models
    13
  5. BenchmarkMathArena BrokenArXiv03/2026
    Deprecated
    Successor of MathArena BrokenArXiv
    Scale
    56 problems
    Primary metric
    Mean judged score
    Applicable models
    13
  6. BenchmarkMathArena BrokenArXiv02/2026
    Deprecated
    Scale
    31 problems
    Primary metric
    Mean judged score
    Applicable models
    13

Family methodology

No family-level methodology recorded.

Across definitions

Tasks
None recorded
Sources
None recorded at family level
Last reviewed
2026-09-22

Applicability does not mean evaluated; benchmark result/model score data is outside this UI contract.