Newest papers

HoloCosm: Beyond Single-Metric Benchmark Compression Across Multiple Metrics

Yupei Lin, Zhongzhan Huang, Junxin Li, Guoming Ling, Shanshan Zhong, Hefeng Wu, Yukai Shi, Liang Lin · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26714 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

A compressão de índice de referência reduz os custos de avaliação selecionando um pequeno conjunto de amostras. Os métodos existentes geralmente guiam a seleção usando uma única métrica, embora os índices de referência muitas vezes avaliem modelos ao longo de múltiplas dimensões. Descobrimos que os rankings de modelo são frequentemente fracos em correlação entre as métricas, indicando que diferentes métricas capturam aspectos distintos do desempenho do modelo em mais de 70\% das comparações cross-metricas. Para resolver este desafio, apresentamos o HoloCosm, que, em conjunto, minimiza o erro de pontuação e reconstrução normalizada em modelos de origem e métricas, enquanto restringe a distorção de classificação. Avaliamos o HoloCosm em taxas de compressão

Machine translation (Meta M2M100); the English original follows.

Abstract

Benchmark compression reduces evaluation cost by selecting a small set of samples. Existing methods typically guide selection using a single metric, although benchmarks often assess models along multiple dimensions. We find that model rankings are often weakly correlated across metrics, indicating that different metrics capture distinct aspects of model performance. Samples selected for one metric do not reliably preserve scores and rankings on others when used for compressed evaluation. At 80\% compression, each of the five single-metric baselines yields higher reconstruction error than equal-size random sampling in over 70\% of cross-metric comparisons. To address this challenge, we present HoloCosm, which jointly minimizes normalized score-reconstruction error across source models and metrics while constraining ranking distortion. We evaluate HoloCosm across multiple compression rates on 15 language, multimodal, VLA, and agent benchmark configurations with 85 component metrics. With 80\% compression, HoloCosm outperforms Random, the strongest baseline overall: average NMAE drops by 35.3\% to 0.236, worst-component NMAE drops by 39.5\% to 0.330, and average Spearman correlation rises from 0.884 to 0.915. In addition, HoloCosm remains competitive with existing baselines in single-metric evaluation. We will release our code and experimental results.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Benchmark (surveying), Metric (unit), Computer science, Compression (physics), Set (abstract data type), Data mining

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange