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
-
GDPval
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
AlpacaEval
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
EQ-Bench
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
HELM AIR-Bench
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
HELM Safety
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
NarrativeQA
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
AdvBench
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
AgentDojo
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
AgentHarm
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
Arena-Hard
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
CTFBench
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
MoralChoice
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
RealToxicityPrompts
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
BOLD
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
FreshQA
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
MT-Bench
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
WeirdML
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
CyberSecEval
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
Generalization Spectrum
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
NeedleBench
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
ETHICS
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
Generalized Zero-Shot Benchmark for Facial Age Estimation
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
InfiniteBench
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
RGB
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
StereoSet
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
CrowS-Pairs
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
HELM Long Context
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
L-Eval
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
LMArena Text Arena
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
MultiHop-RAG
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
-
Needle In A Haystack
No line from its own site or paper places it in Generalization yet; the catalog lists it here.
Next: pick another domain. Tags come from the RL Research daily scrape, canonicalized nightly.