Skip to content

DEV Community

Building a composite benchmark score Series' Articles

Back to Alex Fank's Series

Cover image for Why averaging LLM benchmarks gives the wrong leaderboard

Alex Fank

Oct 5

Why averaging LLM benchmarks gives the wrong leaderboard

#llm #machinelearning #benchmark #datascience

6 min read

Cover image for Fitting one score across 40 sparse benchmarks with item response theory

Alex Fank

Oct 9

Fitting one score across 40 sparse benchmarks with item response theory

#llm #machinelearning #statistics #datascience

10 min read