Skip to content

DEV Community

Leaderboard Forensics Series' Articles

Back to Ward Ed's Series

Cover image for When a 0.4-Point Lead Means Nothing: Reading Open-Model Leaderboard Margins Like a Statistician

Ward Ed

Oct 6

When a 0.4-Point Lead Means Nothing: Reading Open-Model Leaderboard Margins Like a Statistician

#huggingface #llm #benchmark #machinelearning

7 min read

Cover image for Benchmark Contamination 101: How Train/Test Overlap Inflates Leaderboard Scores (and How to Catch It)

Ward Ed

Oct 6

Benchmark Contamination 101: How Train/Test Overlap Inflates Leaderboard Scores (and How to Catch It)

#machinelearning #llm #benchmark #datascience

7 min read

Ward Ed

Oct 12

How to Read a Public Leaderboard Without Fooling Yourself

#ai #machinelearning #benchmark #datascience

3 min read

Cover image for Best Open-Weight LLM in October 2026? Auditing Two Leaderboards Row by Row

Ward Ed

Oct 19

Best Open-Weight LLM in October 2026? Auditing Two Leaderboards Row by Row

#ai #machinelearning #llm #benchmark

12 min read

Cover image for Do 66 Terminal Tasks and a 45% Private Weight Settle AI Rankings? Reading Artificial Analysis Index v4.3

Ward Ed

Oct 20

Do 66 Terminal Tasks and a 45% Private Weight Settle AI Rankings? Reading Artificial Analysis Index v4.3

#ai #machinelearning #llm #benchmark

12 min read

Cover image for GPQA Diamond Has 198 Questions: What the Hugging Face Leaderboard's 115 Rows Can and Cannot Rank

Ward Ed

Oct 21

GPQA Diamond Has 198 Questions: What the Hugging Face Leaderboard's 115 Rows Can and Cannot Rank

#ai #opensource #llm #benchmark

14 min read

Cover image for Reading a Hardware-Equivalence Claim Like an Evaluator: A Checklist Applied to an NPU vs GPU Pretraining Twin

Ward Ed

Oct 26

Reading a Hardware-Equivalence Claim Like an Evaluator: A Checklist Applied to an NPU vs GPU Pretraining Twin

#ai #machinelearning #benchmark #llm

2 min read