Functional Attention: From Pairwise Affinities to Functional Correspondences

Hugging Face Daily Papers 05/29/26, 12:00 AM Papers

operator-learning attention-mechanism transformers function-spaces pde-solving 3d-segmentation deep-learning

Summary

Functional Attention is a novel attention mechanism that reinterprets attention as a functional correspondence between adaptive bases, replacing softmax affinities with structured linear operators inspired by geometric functional maps. The method achieves state-of-the-art performance on operator learning tasks including PDE solving and 3D segmentation while remaining resolution-invariant.

Learning mappings between infinite-dimensional function spaces, or operator learning, is essential for many machine learning applications. Although transformer-based operators are popular, they often rely on token-wise attention. These methods treat continuous fields as discrete tokens and usually ignore the global functional structure. We introduce Functional Attention, which reinterprets attention as a functional correspondence between adaptive bases. Inspired by geometric functional maps, our method replaces softmax affinities with structured linear operators. This yields a compact, generalizable, resolution-invariant representation that explicitly captures global dependencies. Experiments demonstrate that Functional Attention can match state-of-the-art performance in many operator learning tasks, including solving PDEs, 3D segmentation, and regression, while remaining robust to varying discretizations. Project page is available at https://github.com/xjffff/FUNCATTN.

Original Article

Similar Articles

@Phoenixyin13: I think this is a top-notch work in ICML 2026. The attention mechanism of traditional Transformers is essentially point-to-point matching: it cuts input into a bunch of tokens (discrete points), computes similarity between Query and Key, and then weights the Value. In NLP...

X AI KOLs Timeline

Introduces the ICML 2026 paper Functional Attention, which treats functions as first-class citizens and replaces softmax point-to-point similarity with structured linear operators. It addresses issues of discretization, resolution sensitivity, and high computational complexity in traditional Transformers when handling continuous functions. Achieves or surpasses SOTA in tasks like PDE solving and 3D segmentation, and exhibits strong OOD generalization.

Bifocal Attention: Harmonizing Geometric and Spectral Positional Embeddings for Algorithmic Generalization

arXiv cs.CL

This paper introduces Bifocal Attention, which decouples positional encoding into geometric (standard RoPE) and spectral (learnable harmonic operators) modalities to address the 'Spectral Rigidity' of fixed RoPE, improving algorithmic generalization beyond the training window.

Rethinking the Role of Efficient Attention in Hybrid Architectures

arXiv cs.CL

This paper systematically analyzes the role of efficient attention modules in hybrid language model architectures, finding that different designs converge in long-context performance under sufficient training, and that long-range retrieval is primarily carried by full attention while efficient attention shapes the optimization trajectory, revealing a 'Large-Window Laziness' phenomenon.

@tetsuoai: Attention is a lookup. Each token builds a query, compares it against every key in the sequence, and pulls value vector…

X AI KOLs Timeline

Explains attention in transformers as a lookup operation where each token builds a query, compares against keys, and retrieves weighted value vectors, with a video covering the full pipeline.

The Routing and Filtering Structure of Attention

arXiv cs.LG

The paper decomposes the attention interaction matrix into routing (skew-symmetric) and filtering (symmetric) components, introducing S-D attention to disentangle them. It reveals a spectral cascade in routing that predicts where attention can be simplified, achieving significant parameter reduction with minimal perplexity loss.