COMPASS:訓練前評估模型配對是否適合對齊或合併
To Align or Not To Align: Check Your COMPASS Before You Train
COMPASS 在模型對齊或合併前評估配對相容性,協助排序候選配對及配置資源。它以編碼器間的 k-NN 重疊度及訓練範式可達的重疊上限,作為兩項可低成本計算的指標。在 72 項跨模態對齊實驗及 392 項任務算術模型合併中,COMPASS 與下游成效的平均 Pearson 相關係數分別為 0.91 及 0.70。
Published on Oct 13
Authors:
,
Abstract
COMPASS estimates model-pair compatibility before alignment or merging by measuring local neighborhood overlap and paradigm limits, guiding resource allocation across cross-modal and merging tasks.
Combining pretrained models is a common strategy for building AI systems, whether by aligning a vision encoder with a language model or merging fine-tuned checkpoints. However, finding compatible pairs typically requires costly empirical search, as compatibility is known only after alignment or merging. We propose COMPASS (COMPatibility ASsessment Score), a theoretically motivated diagnostic that estimates whether a model pair will be compatible prior to alignment or merging. Building on the observation that representational convergence is most robust at the level of local neighborhoods, COMPASS decomposes model compatibility into two cheap-to-compute axes: a pair-specific ease, defined as the k-NN overlap between the two encoders, and a paradigm-level ceiling, capturing the maximum overlap their training paradigms allow. Together, these two axes provide a ranking of candidate model pairs and prescribe how to act on each pair: (i) prioritize pairs that already agree and whose paradigms are compatible, (ii) invest additional compute on pairs with compatible paradigms but low initial agreement, (iii) switch the pretraining paradigm when it cannot support sufficient agreement, or (iv) deprioritize pairs that score poorly on both axes. We apply COMPASS to 72 cross-modal alignment experiments and 392 task-arithmetic model merges and show that compatibility estimates by COMPASS strongly correlate with downstream success, achieving average Pearson correlation of 0.91 on cross-modal alignment tasks and 0.70 on model merging.
View arXiv page View PDF Project page GitHub 0 Add to collection
Get this paper in your agent:
hf papers read 2610.18638
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.18638 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.18638 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.18638 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co