Claude suddenly stopped cheating – Drone Bench | Hacker News - But Better

Claude suddenly stopped cheating – Drone Bench twitter.com

1 points•infogulch•1 day ago•1 comment•

#bench #claude #drone #suddenly #cheating #stopped #claude suddenly #suddenly stopped

1 comment

infogulch1 day ago

This X thread highlights how opus-5.5 reversed the trend of Claude models found to be cheating on Drone-Bench.

Source: https://andonlabs.com/evals/drone-bench#:~:text=Cheating%20s...

Read the full thread on Hacker News →

Related stories

Military milestone: Ukrainian naval drone sinks Russian kamikaze drone boatarstechnica.com
Ars Technica · 0 points · 9 days ago
Does this mean it's fine to use Claude subscriptions with third party harnesses?
Hacker News · 4 points · 13 days ago
Getting the most out of Opus 5.5 in Claude and Claude Code / claude.devclaude.dev
Hacker News · 2 points · 7 days ago
Show HN: use-citations – Claude Code skill for citation backed answersgithub.com
Hacker News · 1 points · 8 days ago
Show HN: Migrating from Claude Code CLI to Claude Desktopgithub.com
Hacker News · 1 points · 11 days ago
Getting the most out of Opus 5.5 in Claude and Claude Codeclaude.dev
Hacker News · 2 points · 8 days ago