Labs Benchmark LLMs Against Leaked Open Evals, Effectively Testing on Train Set
fraserpricee · x · 2026-10-18
Developer @fraserpricee mocks AI labs for continuing to benchmark their LLMs on openly available evals (or compromised/leaked closed ones), arguing this is effectively testing on the training set. The take sparked a debate with @evijit, who pushed back that open-source evals aren't the problem—contamination testing should be solved as a technical issue—and raised follow-up concerns: how can third parties measure eval quality, who pays for the compute, and what about small OSS labs and community developers.
More from Models
- OpenAI's Decisions API hits GA: near real-time routing up to 10x faster than GPT-6 Luna — OpenAIDevs · 2026-10-23
- DeepLearning.AI weekly: GPT-6.1 hits 97.9% byte parity on COBOL migration — DeepLearningAI · 2026-10-23
- Marker's lite mode scores 75.1 vs 42.3 for Azure Read on OmniParseBench — VikParuchuri · 2026-10-23
- Datalab launches lite OCR at $0.50 per 1,000 pages, 20-30x cheaper than Azure, Google and AWS layout APIs — VikParuchuri · 2026-10-23
- LiveNerf's 29-day data finds no nerfing of Claude Opus 5.5 — TheOnlyVibemaster · 2026-10-23
- PHP Foundation reveals Anthropic's Claude cyber model scanned PHP, validated with OpenAI's Daybreak Red — saibharadwaj · 2026-10-23