No Triangulation Without Representation: Generalization in Topological Deep Learning
cs.LG, math.AT
Submitted: 2026-05-07
Updated: 2026-11-09
Comments: Accepted at the 40th Conference on Neural Information Processing Systems (NeurIPS) 2026. Our code is available at https://github.com/aidos-lab/mantra-pp-benchmarks
License: http://creativecommons.org/licenses/by/4.0/
The gist: Despite an ever-increasing interest in topological deep learning models that target higher-order datasets, there is no consensus on how to evaluate such models.
Terminology
Abstract
Despite an ever-increasing interest in topological deep learning models that target higher-order datasets, there is no consensus on how to evaluate such models. This is exacerbated by the fact that topological objects permit operations, such as structural refinements, that are not appropriate for graph data. In this work, we extend MANTRA, a benchmark dataset containing manifold triangulations, to a larger class of manifolds with more diverse homeomorphism types. We show that, contrary to prior claims, graph neural networks can saturate the benchmark, while higher-order message passing methods lag behind. However, we find that this is contingent on the right representation and feature assignment, emphasizing their importance in baseline models. Hence, we present a novel evaluation protocol based on representational diversity and triangulation refinement. Even when including refinement-like changes in the training data, existing models fail to generalize beyond the combinatorial structure of the data. This points towards a research gap in developing models that understand topological structure independent of scale. Our work thus provides the necessary scaffolding to evaluate future models and enable the development of topology-aware inductive biases.
Sources
- Attending to Topological Spaces: The Cellular Transformer
- Residual Gated Graph ConvNets
- HOPSE: Scalable Higher-Order Positional and Structural Encoder for Combinatorial Representations
- Graphs are maximally expressive for higher-order interactions
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis