Search papers, labs, and topics across Lattice.
This paper investigates the reliability of training-free node classification using tabular foundation models (TFMs) by evaluating conformal coverage and prediction set size. The study demonstrates that the training-free TabICL TFM achieves lower expected calibration error than a trained GCN with temperature scaling on most tested graphs. Furthermore, the proposed homophily-gated diffusion score, HG-DAPS, reduces prediction set size on homophilous graphs while maintaining conformal validity.
Training-free TFMs can outperform trained GCNs in calibration for node classification, and a novel homophily-gated approach significantly shrinks prediction sets without sacrificing validity.
Tabular foundation models (TFMs) can classify the nodes of a graph without training on it, by reading node and neighborhood features as table rows next to labeled context rows. Work in this line reports predictive performance, not conformal coverage or prediction-set size. To our knowledge, we give the first reliability study of the setting, with TabICL as the TFM and half of each graph as labeled context. As for any predictor fixed before calibration, a frozen in-context predictor makes split conformal prediction exactly valid in finite samples, with no training, validation fold, or tuning on the target graph. An audit across ten graphs then shows that the training-free TabICL posterior has lower expected calibration error (ECE) than GCN with temperature scaling (GCN+TS) on nine of them. Its mean ECE over the ten graphs is 0.019, about 35 percent below the 0.029 of GCN+TS. We also introduce HG-DAPS, a training-free diffusion score whose homophily gate reads only the in-context labels, so the guarantee still holds. Relative to adaptive prediction sets (APS), it reduces mean set size by 5.8 to 17.1 percent on six homophilous graphs and changes it by under 1 percent on four heterophilous ones. On two binary, class-imbalanced graphs, a pre-registered trap case shows that gating on raw rather than adjusted homophily lowers coverage among low-homophily nodes by 0.27 and 0.12. Marginal coverage stays at the nominal 0.90 and masks this drop.
