arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.06205cs.LGcs.AI
收藏

多向表示对齐

Multi-Way Representation Alignment

Akshit Achara, Tatiana Gaintseva, Mateo Mahaut, Pritish Chakraborty, Viktor Stenby Johansson, Melih Barsbey, Emanuele Rodolà, Donato Crisostomi

AI总结:

研究M≥3模型的表示对齐,用广义普罗克汝斯忒斯分析构建共享正交空间,指出严格等距对齐对检索非最优,提出几何校正普罗克汝斯忒斯对齐,实验表明其提升任意到任意检索并保留实用共享参考空间。

摘要:

柏拉图式表示假设表明独立训练的神经网络收敛到越来越相似的潜在空间。然而,当前映射这些表示的策略本质上是成对的,与模型数量成二次方缩放且无法产生一致的全局参考。本文研究M≥3个模型的对齐。首先采用广义普罗克汝斯忒斯分析构建共享正交空间以保留模型拼接等任务所需的内部几何结构。然后表明严格等距对齐对检索非最优,其中典型的最大化一致性方法如典型相关分析占主导。为弥合这一差距,最终提出几何校正普罗克汝斯忒斯对齐,它基于稳健的广义普罗克汝斯忒斯分析构建空间并对方向失配进行事后校正。大量实验表明几何校正普罗克汝斯忒斯对齐在保留实用共享参考空间的同时持续改善任意到任意检索。

英文摘要:

The Platonic Representation Hypothesis suggests that independently trained neural networks converge to increasingly similar latent spaces. However, current strategies for mapping these representations are inherently pairwise, scaling quadratically with the number of models and failing to yield a consistent global reference. In this paper, we study the alignment of $M \ge 3$ models. We first adapt Generalized Procrustes Analysis (GPA) to construct a shared orthogonal universe that preserves the internal geometry essential for tasks like model stitching. We then show that strict isometric alignment is suboptimal for retrieval, where agreement-maximizing methods like Canonical Correlation Analysis (CCA) typically prevail. To bridge this gap, we finally propose Geometry-Corrected Procrustes Alignment (GCPA), which establishes a robust GPA-based universe followed by a post-hoc correction for directional mismatch. Extensive experiments demonstrate that GCPA consistently improves any-to-any retrieval while retaining a practical shared reference space.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑