Benchmark or dataset
19 entities, most discussed first
PubMed
9 episodes
Online database of biomedical literature abstracts maintained by US National Library of Medicine
Human Development Index
3 episodes
Composite index ranking countries by life expectancy, education, and income
MMLU
3 episodes
Multi-task benchmark testing language understanding across academic subjects
ANSI/TIA-568
2 episodes
Telecommunications cabling standard specifying wire pairing, termination and performance requirements
BLEU
2 episodes
Automatic metric for scoring machine translation quality against reference texts
Common Crawl
2 episodes
Nonprofit web crawl archive widely used as AI training data
CPAN
2 episodes
Perl package repository
FineWeb
2 episodes
Large open web text dataset derived from Common Crawl for language model pretraining
HotpotQA
2 episodes
Question-answering dataset requiring multi-hop reasoning over Wikipedia passages
SimpleQA
2 episodes
Factual question-answering benchmark dataset released by OpenAI
SWE-bench
2 episodes
Software engineering benchmark evaluating AI models on resolving real GitHub issues
Terminal-Bench
2 episodes
Benchmark evaluating AI agents on command-line terminal tasks
TIOBE index
2 episodes
Monthly ranking of programming language popularity based on search engine hits
ABFO Number 2 Standard Reference Scale
1 episode
Standardised scale used in forensic bite mark analysis reference photography
Cohen's d
1 episode
Standardised effect size measure for the difference between two group means
English Speller Database
1 episode
English word list and spelling database maintained by Kevin Atkinson
PISA
1 episode
Triennial OECD international assessment of fifteen-year-olds' reading, maths and science
SCOWL
1 episode
Open-source English spell-checking word list used in dictionaries
Tokengeist
1 episode
Microsoft method for multi-turn context attribution in language models