Home▸AI & Machine Learning▸Benchmark Contamination Detector

Benchmark Contamination Detector

Interactive 3D lab: watch how leaked benchmark items inflate a model's reported accuracy, and how an embedding-similarity threshold trades detection recall against false positives when trying to catch the leakage.

AI & Machine Learning3DModerate60 FPS📱 Mobile-adapted⇄ 2D version

Every benchmark score for a large model carries a hidden question: how much of that score comes from the model actually reasoning, versus having simply seen the answer during pretraining? This simulator makes benchmark contamination physical. A ring of glowing spheres represents individual eval questions, each lifted to a height equal to its embedding similarity against a training-corpus cloud; a translucent plane is the similarity threshold a decontamination filter would use to flag and discard suspect items before re-scoring. Dial the true contamination rate, the model's genuine skill, and the detector's threshold, and watch the reported accuracy separate from the decontaminated accuracy in real time — along with the recall/false-positive trade-off that governs every real contamination audit.

⚙ Under the hood

Watch leaked benchmark items inflate a model's reported accuracy, and see how an embedding-similarity threshold trades detection recall against false positives when trying to catch the leakage.

LLM evaluationdata leakagebenchmark contaminationembedding similaritymodel evaluation

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install