Don't let the training team see your internal benchmarks — just return a score
xeophon · x · 2026-11-09
In a discussion with Megan Risdal and Hamel Husain, xeophon argues the people doing training should not have access to internal benchmarks or even know what's inside — training just gets an endpoint and receives a score back, to avoid goodharting. llmwizard adds that internal benchmark suites should also be hill-climbed judiciously.
Related event: AI Benchmark Leaderboards Undermined by Goodharting, Experts Say(4 posts)→
More from coding & agent
- Simon Willison hides AI PR descriptions in details/summary tags — Simon Willison · 2026-11-09
- Team's AI agent fixes 40 of 41 user-reported issues in a single weekend — julianweisser · 2026-11-09
- HEELO Harness Framework Lifts Terminal-Bench Resolution from 37.3% to 55.5% — Haibo Jin · 2026-11-09
- NVIDIA's AgentBaton: Well-Timed Strong-to-Weak Model Handoffs Slash Agent Costs — nvidia · 2026-11-09
- FrontierCS 2 update: Claude Sonnet 5.5 beats reference on 50% of tasks, reward hacks published — ajratner · 2026-11-09
- "Being technical" now means having the patience to ask Claude until you get it — brandon_galang · 2026-11-09