English(EN) Contamination Inflates Scores but Rarely Reorders Large Language Model Leaderboards

LLM基准测试污染推高分数但很少重排排行榜

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-04 04:00

arXiv上的一篇新论文研究了大型语言模型中的基准测试污染，区分了分数膨胀和排行榜重排。研究发现，虽然污染确实会推高绝对分数，但很少改变模型在排行榜上的排名。该研究提出了一种通过比较原始测试项目和释义版本来审计污染的方法，并建议排行榜应报告释义控制的排名以及置信区间。 AI

影响提供了一种审计LLM基准测试污染的方法，提高了模型评估的可靠性。

排序理由研究论文分析LLM基准测试污染。[lever_c_demoted from research: ic=1 ai=1.0]

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 16 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 研究论文分析LLM基准测试污染。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

arXiv cs.CL TIER_1 English(EN) · Xingyao Xiao (Stanford University), Yihong Cheng (City University of Macau) · 2026-09-04 04:00

污染夸大评分但很少重排大型语言模型排行榜

arXiv:2609.02899v1 Announce Type: new Abstract: 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: wheth…