News for hackers. Stuff that matters.

A comment on Problems in LLM Benchmarking and Evaluation.

Re:Problems in LLM Benchmarking and Evaluation by tianlong on Friday August 15, 2025 @11:50AM (#44913868)
What's the TLDR about how to solve that benchmark problem?
[ Parent | on HN ]
Re:Problems in LLM Benchmarking and Evaluation by upperhalfplane on Friday August 15, 2025 @11:54AM (#44913939)
TLDR: games are a good way to go.
[ Parent | on HN ]