Directory label

Generalization: AI benchmarks

A label other catalogs put on entries, not one of the domains on our map. On the map, each entry here sits in the domain its own site or paper leads with.

AI benchmarks
33
Shares entries with
0 categories

Snapshot 2026-11-12. How categories are built · Where the data comes from

In short

What are the main generalization benchmarks?
GDPval, AlpacaEval, EQ-Bench, HELM AIR-Bench, HELM Safety and NarrativeQA, among 31 confirmed below.

Only benchmarks use the Generalization tag: the RL environment directories file their vendors under other categories. Browse all RL environments.

AI benchmarks 33

  1. GDPvalOpenAI
  2. AlpacaEval
  3. EQ-Bench
  4. HELM AIR-BenchStanford Center for Research on Foundation Models, Virginia Tech, other universities
  5. HELM SafetyStanford Center for Research on Foundation Models
  6. NarrativeQA
  7. AdvBench
  8. AgentDojoInvariant Labs, ETH Zurich
  9. AgentHarm
  10. Arena-HardLMArena
  11. CTFBench
  12. MoralChoice
  13. RealToxicityPromptsAi2
  14. BOLD
  15. FreshQA
  16. MT-BenchLMSYS
  17. WeirdML
  18. CyberSecEvalMeta
  19. Generalization Spectrum
  20. NeedleBenchOpenCompass
  21. ETHICSDan Hendrycks, Collin Burns, Steven Basart, Andrew Critch
  22. Generalized Zero-Shot Benchmark for Facial Age Estimation
  23. InfiniteBench
  24. RGB
  25. StereoSet
  26. CrowS-PairsNew York University
  27. HELM Long ContextStanford Center for Research on Foundation Models
  28. L-EvalOpenLMLab
  29. LMArena Text ArenaLMSYS
  30. MultiHop-RAG
  31. Needle In A Haystack
2 not yet confirmedListed in the catalog; no line from its own site or paper confirms it yet.
  1. RAGAS
  2. HH-RLHF

Next: pick another domain. Tags come from the RL Research daily scrape, canonicalized nightly.