H
Howardism
Howardism · Vol. 03Plate I · No. 01

Benchmarks, tagged.

Notes19TagBenchmarksOldest7 Jun 2026Newest27 Aug 2026

Every article tagged benchmarks, newest first.

  1. C01
    Skill Lift
    EvaluationAgent SkillsAblation+3
    · 15′
  2. C02
    GDPval Benchmark
    EvaluationBenchmarksEconomic Impact+1
    · 12′
  3. C03
    Aggregate Cancellation
    Evaluation MethodologyConstruct ValidityBenchmarks+1
    · 9′
  4. C04
    Frontier AI Standards Body
    GovernanceAI PolicyFrontier Labs+2
    · 20′
  5. C05
    Expenditure Horizon
    BenchmarksCapability EvaluationEvaluation Methodology+3
    · 22′
  6. C06
    Agent-Authored Harness Optimization
    Agent EngineeringHarnessEvaluation+2
    · 82′
  7. C07
    Knowledge-Centric Self-Improvement
    Agent EngineeringKnowledge ManagementRecursive Self Improvement+2
    · 33′
  8. C08
    Orchestration-Plan Simulation
    BenchmarksEvaluation MethodologyMulti Agent+2
    · 24′
  9. C09
    Benchmark Contamination and Decontamination
    LLM ArchitectureCapability EvaluationBenchmarks+4
    · 17′
  10. C10
    Benchmark Score Redundancy
    LLM ArchitectureCapability EvaluationBenchmarks+3
    · 29′
  11. C11
    Measuring Beyond Accuracy Saturation
    LLM ArchitectureCapability EvaluationBenchmarks+3
    · 42′
  12. C12
    Reference-Free Judge Over-Crediting
    EvaluationLLM As A JudgeBenchmarks+3
    · 41′
  13. C13
    LLM-Judge Validation
    EvaluationLLM As A JudgeBenchmarks+2
    · 54′
  14. C14
    Compute-Controlled Benchmarking
    LLM ArchitectureCapability EvaluationBenchmarks+2
    · 41′
  15. C15
    DRACO Benchmark
    BenchmarksCapability EvaluationDeep Research+1
    · 10′
  16. C16
    LLM-as-a-Judge
    EvaluationLLM As A JudgeBenchmarks+1
    · 36′
  17. C17
    Production-Sourced Evaluation
    EvaluationBenchmarksData Provenance+1
    · 22′
  18. E18
    METR
    EntityOrgAI Evaluation+1
    · 17′
  19. C19
    Task Time-Horizon Scaling
    LLM ArchitectureCapability EvaluationBenchmarks+1
    · 23′