# Grables: When Graph Structure Helps Tabular Prediction

arXiv: https://arxiv.org/abs/2602.03945 · HTML: https://mlsanity.com/paper/2602.03945/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 33 references that are in the map.
- First posted on arXiv 3 Feb 2026.

## Builds on (most cited first, 21 of 21)

- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [Neural Message Passing for Quantum Chemistry](https://mlsanity.com/paper/1704.01212/) (2017)
- [Modeling Relational Data with Graph Convolutional Networks](https://mlsanity.com/paper/1703.06103/) (2017)
- [Revisiting Deep Learning Models for Tabular Data](https://mlsanity.com/paper/2106.11959/) (2021)
- [TabTransformer: Tabular Data Modeling Using Contextual Embeddings](https://mlsanity.com/paper/2012.06678/) (2020)
- [AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data](https://mlsanity.com/paper/2003.06505/) (2020)
- [TabArena: A Living Benchmark for Machine Learning on Tabular Data](https://mlsanity.com/paper/2506.16791/) (2025)
- [TabPFN-2.5: Advancing the State of the Art in Tabular Foundation Models](https://mlsanity.com/paper/2511.08667/) (2025)
- [TabDPT: Scaling Tabular Foundation Models on Real Data](https://mlsanity.com/paper/2410.18164/) (2024)
- [TabM: Advancing Tabular Deep Learning with Parameter-Efficient Ensembling](https://mlsanity.com/paper/2410.24210/) (2024)
- [RelBench: A Benchmark for Deep Learning on Relational Databases](https://mlsanity.com/paper/2407.20060/) (2024)
- [Better by Default: Strong Pre-Tuned MLPs and Boosted Trees on Tabular Data](https://mlsanity.com/paper/2407.04491/) (2024)
- [CARTE: Pretraining and Transfer for Tabular Learning](https://mlsanity.com/paper/2402.16785/) (2024)
- [word2vec, node2vec, graph2vec, X2vec: Towards a Theory of Vector Embeddings of Structured Data](https://mlsanity.com/paper/2003.12590/) (2020)
- [Relational Graph Transformer](https://mlsanity.com/paper/2505.10960/) (2025)
- [GraphLand: Evaluating Graph Machine Learning Models on Diverse Industrial Data](https://mlsanity.com/paper/2409.14500/) (2024)
- [Table Foundation Models: on knowledge pre-training for tabular learning](https://mlsanity.com/paper/2505.14415/) (2025)
- [Graph Neural Networks for Tabular Data Learning: A Survey with Taxonomy and Directions](https://mlsanity.com/paper/2401.02143/) (2024)
- [From Features to Structure: Task-Aware Graph Construction for Relational and Tabular Learning with GNNs](https://mlsanity.com/paper/2506.02243/) (2025)
- [Fixed Aggregation Features Can Rival GNNs](https://mlsanity.com/paper/2601.19449/) (2026)

## Built on by (most cited first, 2 of 2)

- [Tabular Foundation Models for Discrete Choice Estimation](https://mlsanity.com/paper/2607.13314/) (2026)
- [AutoGrable: What Is a Good Graph for a Table?](https://mlsanity.com/paper/2608.11431/) (2026)

## Abstract (arXiv)

Simple row-wise predictors that score each row independently remain competitive with far more expressive architectures on tabular benchmarks, which are i.i.d. by construction. They cannot, however, represent natural targets on transactional, temporal, and relational tables, where labels depend on other rows. We show that row-wise prediction provably rules out targets driven by global counts, overlaps, and relational patterns. To make ``using structure'' precise across architectures, we introduce grables: a modular interface that separates how a table is lifted to a graph (constructor) from how predictions are computed on that graph (node predictor), placing gradient-boosted trees, neighbourhood aggregation, transformer-style tabular models, and relational GNNs on a single axis and pinpointing where expressive power comes from. The framework yields a hierarchy that answers when structure is needed: row-local models suffice for extension-invariant targets, lightweight neighbourhood aggregation captures shallow inter-row signals, and explicit graph construction is needed for compositional patterns that fixed aggregations cannot express without exponential feature blow-up. Experiments on synthetic, transactional, and relational data confirm the predicted separations, but also show that benchmark performance is a poor proxy for structural capability: row-local models score well on tasks they provably cannot represent, and a stress-test that suppresses row-local cues is needed to expose the gap. Conversely, access to cross-row context does not imply a model uses it. Feeding explicitly extracted inter-row structure to strong tabular learners yields gains exactly where the hierarchy predicts. The practical recipe is to add structure incrementally, and only where the target is extension-sensitive.