Tag

#Contamination

Contamination on The Latent Daily: 1 story and 1 other resource. Latest: How AI leaderboards rank models: votes, contamination and the ways rankings get gamed.

2 items

A heatmap of chess game outcomes plotted by the Elo ratings of the two players

Benchmarks & Leaderboards

How AI leaderboards rank models: votes, contamination and the ways rankings get gamed

The three main kinds of AI leaderboard, how arena-style voting turns clicks into ratings, and where contamination and selective reporting distort the order.

thelatentdaily.com Editorial TeamOct 256 min read

Benchmarks tagged #Contamination

View all

Benchmarks

LiveCodeBench

A continuously updated coding benchmark built from new LeetCode, AtCoder and Codeforces contest problems, dated so models can be tested on problems released after their training.