Metraipolis › archive › dc_b2def2d716a7
Contamination Inflates Scores but Rarely Reorders Large Language Model Leaderboards — arXiv:2609.02899
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.02899
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T01:37:18.699Z
- SHA-256 of the archived text
e5cde2b2af99e3beadef2e718f5059d49c506eb4861dea131e9d5abf0878b38b- Stable URL of this page
https://metraipolis.com/documento/dc_b2def2d716a7
How to cite
Contamination Inflates Scores but Rarely Reorders Large Language Model Leaderboards — arXiv:2609.02899 — arXiv (en), https://arxiv.org/abs/2609.02899 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b2def2d716a7 (Metraipolis archive dc_b2def2d716a7, sha256 e5cde2b2af99e3beadef2e718f5059d49c506eb4861dea131e9d5abf0878b38b)
BibTeX
@misc{dcb2def2d716a7,
title = {Contamination Inflates Scores but Rarely Reorders Large Language Model Leaderboards — arXiv:2609.02899},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.02899},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b2def2d716a7; sha256 e5cde2b2af99e3beadef2e718f5059d49c506eb4861dea131e9d5abf0878b38b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.02899. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Benchmark contamination, the leakage of test items into training data, is widely described as a threat to the reliability of large language model (LLM) leaderboards. We argue that this concern conflates two distinct questions: whether contamination inflates absolute scores, and whether it reorders the ranking of models. We recast contamination as a violation of anchor-item invariance and measure it through the differential functioning of original versus semantically equivalent paraphrased items, a within-item contrast that holds the measured skill fixed and isolates memorization from capability. Using per-instance responses from 47 publicly released models and 74 models finetuned with a known dose of contamination, across four benchmarks (ARC, GSM8K, HellaSwag, MMLU), we first calibrate the measure against ground truth: it recovers injected contamination dose-responsively (a corrected effect of +0.187 accuracy points for test-set leakage) and never flags a negative-control model trained only on the legitimate training split (-0.012). We then quantify leaderboard impact: the rank correlation between a standard leaderboard and a paraphrase-controlled leaderboard is 0.997, and a sensitivity analysis shows that the observed differential contamination is far below the level needed to move rankings, with only 3 of 188 model-by-benchmark cases showing differential contamination corroborated across two references. Contamination among these public models is therefore largely uniform: it inflates absolute scores without reordering the leaderboard, and ranking distortion requires the rare case of differential contamination. We provide a calibrated invariance audit, released as a reference implementation, and recommend that leaderboards report paraphrase-controlled rankings alongside confidence intervals. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.CL). Categorie arXiv: Computation and Language (cs.CL) ; Applications (stat.AP); Methodology (stat.ME). Pagina del lavoro: https://arxiv.org/abs/2609.02899 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T01:11:43.489Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).