Shaping the Prior: How Synthetic Task Distributions Determine Tabular Foundation Model Quality

• URL: http://arxiv.org/abs/2605.18971v1
• Date: Mon, 18 May 2026 18:00:45 GMT
• Title: Shaping the Prior: How Synthetic Task Distributions Determine Tabular Foundation Model Quality
• Abstract summary: We introduce O'Prior, a compositional realism prior built around four coupled components.<n>O'Prior yields consistent and substantial improvements in accuracy and robustness across real benchmarks.
• Score: 0.6670498055582529
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: What determines the quality of a tabular foundation model? Unlike language or vision, tabular foundation models acquire their inductive biases almost entirely from synthetic pretraining distributions, yet the design of these distributions remains poorly understood. Standard synthetic priors are too well-behaved: they omit the irregularities and failure modes that determine deployment robustness. We introduce O'Prior, a compositional realism prior built around four coupled components: a hierarchical SCM meta-generator spanning diverse functional families; a modular realism engine covering heterogeneous marginals, missingness, and target transforms; an explicit stress module injecting confounding and support-query mismatch; and a curriculum-governed, leakage-safe generation protocol. To isolate prior design as the scientific variable, we hold architecture, optimizer, and compute budget fixed and vary only the synthetic task distribution. O'Prior yields consistent and substantial improvements in downstream accuracy and robustness across real tabular benchmarks, with gains concentrated in regimes characterized by distributional irregularities. Ablations confirm that mechanism diversity, realism composition, and shift-aware stress each contribute independently, their effects are not interchangeable. These results establish synthetic prior construction as a first-order and largely overlooked determinant of tabular foundation model quality

Related papers

• Position: It's Time to Optimize LLMs for Self-Consistency [50.68587056599734]
  Many important failures persist in language model pre- and post-training pipelines.<n>We argue that these failures arise from a modeling assumption permeating all aspects of the pipeline.<n>We propose self-consistency as a framework for understanding these failures.
  arXiv  Detail & Related papers  (2026-07-31T19:16:53Z)
• Stationary Robust Mean-Field Games under Model Mismatches [3.053409604085541]
  This paper develops an infinite-horizon, stationary mean-field game framework.<n>It incorporates distributional model uncertainty directly into the population-coupled dynamics.<n>Under a contractive robust-dynamics regime, we obtain explicit non-asymptotic error bounds.
  arXiv  Detail & Related papers  (2026-06-21T16:29:25Z)
• Traits Run Deeper: Trait-Specific Asymmetric Fusion for Personality Assessment [53.50824432699408]
  Traits Run Deeper is a novel personality assessment framework.<n>It consists of three components: the Multimodal Foundation Representation (MFR), the Trait-Specific Modality Fusion (TSMF) and the Distribution-Calibrated Personality Regression (DCPR)
  arXiv  Detail & Related papers  (2026-06-09T06:38:36Z)
• Compositional Generalization in Autoregressive Models via Logit Composition [5.3804728961007315]
  We introduce a new and principled composition strategy for autoregressive systems.<n>Under a factorized-conditionals assumption, we show that the resulting composition is projective.<n>We also show that composition preserves length-generalizing behavior when the factorization assumptions and component guarantees hold uniformly at the target length.
  arXiv  Detail & Related papers  (2026-05-27T11:00:26Z)
• A Mechanistic Study of Tabular Foundation Models [12.26738662548555]
  Tabular foundation models converge in accuracy across a range of classification and regression tasks.<n>This raises questions a leaderboard cannot answer.<n>We characterize all three: (i) whether the models execute the same in-context algorithm, (ii) where row, column, and class-permutation invariances originate, and (iii) how robust they are under perturbations engineered against the inferred mechanism.
  arXiv  Detail & Related papers  (2026-05-20T15:23:16Z)
• OrthoFormer: Instrumental Variable Estimation in Transformer Hidden States via Neural Control Functions [0.0]
  Transformer architectures excel at sequential modeling yet remain fundamentally limited by correlational learning.<n>We propose OrthoFormer, a causally grounded architecture that embeds instrumental variable estimation directly into Transformer blocks via neural control functions.
  arXiv  Detail & Related papers  (2026-03-08T03:05:16Z)
• SpanNorm: Reconciling Training Stability and Performance in Deep Transformers [55.100133502295996]
  We propose SpanNorm, a novel technique designed to resolve the dilemma by integrating the strengths of both paradigms.<n>We provide a theoretical analysis demonstrating that SpanNorm, combined with a principled scaling strategy, maintains bounded signal variance throughout the network.<n> Empirically, SpanNorm consistently outperforms standard normalization schemes in both dense and Mixture-of-Experts (MoE) scenarios.
  arXiv  Detail & Related papers  (2026-01-30T05:21:57Z)
• Making Foundation Models Probabilistic via Singular Value Ensembles [56.4174499669573]
  Foundation models have become a dominant paradigm in machine learning, achieving remarkable performance across diverse tasks through large-scale pretraining.<n>Standard approach to quantifying uncertainty, training an ensemble of independent models, incurs prohibitive computational costs that scale linearly with ensemble size.<n>We propose Singular Value Ensemble (SVE), a parameter-efficient implicit ensemble method that builds on a simple, but powerful core assumption.<n>We show that SVE uncertainty quantification achieves comparable to explicit deep ensembles while increasing the parameter count of the base model by less than 1%.
  arXiv  Detail & Related papers  (2026-01-29T18:07:18Z)
• Decoupling the Effect of Chain-of-Thought Reasoning: A Human Label Variation Perspective [60.45433515408158]
  We show that long Chain-of-Thought (CoT) serves as a decisive decision-maker for the top option but fails to function as a granular distribution calibrator for ambiguous tasks.<n>We observe a distinct "decoupled mechanism": while CoT improves distributional alignment, final accuracy is dictated by CoT content.
  arXiv  Detail & Related papers  (2026-01-06T16:26:40Z)
• The Reasoning-Creativity Trade-off: Toward Creativity-Driven Problem Solving [57.652356955571065]
  State-of-the-art large language model (LLM) pipelines rely on bootstrapped reasoning loops.<n>We analyze how this design choice is sensitive to the collapse of the model's distribution over reasoning paths.<n>We introduce Distributional Creative Reasoning (DCR), a unified variational objective that casts training as gradient flow through probability measures on solution traces.
  arXiv  Detail & Related papers  (2026-01-02T17:10:31Z)
• Dictionary-Transform Generative Adversarial Networks [11.62669179647184]
  We introduce emphDictionary-Transform Generative Adversarial Networks (DT-GAN)<n> DT-GAN is a model-based adversarial framework in which the generator is a sparse synthesis dictionary and the discriminator is an analysis transform acting as an energy model.<n>We show that the DT-GAN adversarial game is well posed and admits at least one Nash equilibrium.
  arXiv  Detail & Related papers  (2025-12-25T13:41:14Z)
• Uncertainty-driven Embedding Convolution [16.523816971857787]
  We propose Uncertainty-driven Embedding Convolution (UEC)<n>UEC transforms deterministic embeddings into probabilistic ones in a post-hoc manner.<n>It then computes adaptive ensemble weights based on embedding uncertainty.
  arXiv  Detail & Related papers  (2025-07-28T11:15:25Z)
• Diagnosing and Rectifying Fake OOD Invariance: A Restructured Causal Approach [51.012396632595554]
  Invariant representation learning (IRL) encourages the prediction from invariant causal features to labels de-confounded from the environments. Recent theoretical results verified that some causal features recovered by IRLs merely pretend domain-invariantly in the training environments but fail in unseen domains. We develop an approach based on conditional mutual information with respect to RS-SCM, then rigorously rectify the spurious and fake invariant effects.
  arXiv  Detail & Related papers  (2023-12-15T12:58:05Z)
• Enhancing Robustness of Foundation Model Representations under Provenance-related Distribution Shifts [8.298173603769063]
  We examine the stability of models based on foundation models under distribution shift. We focus on confounding by provenance, a form of distribution shift that emerges in the context of multi-institutional datasets. Results indicate that while foundation models do show some out-of-the-box robustness to confounding-by-provenance related distribution shifts, this can be improved through adjustment.
  arXiv  Detail & Related papers  (2023-12-09T02:02:45Z)
• Neural Ensemble Search for Uncertainty Estimation and Dataset Shift [67.57720300323928]
  Ensembles of neural networks achieve superior performance compared to stand-alone networks in terms of accuracy, uncertainty calibration and robustness to dataset shift. We propose two methods for automatically constructing ensembles with emphvarying architectures. We show that the resulting ensembles outperform deep ensembles not only in terms of accuracy but also uncertainty calibration and robustness to dataset shift.
  arXiv  Detail & Related papers  (2020-06-15T17:38:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.