Position: Foundation Models for Tabular Data within Systemic Contexts Need Grounding

TL;DR

This paper emphasizes the importance of grounding foundation models for tabular data in operational context, introducing Semantically Linked Tables and a new training paradigm to improve their practical understanding and application.

Contribution

It proposes a novel model class, FMSLT, with dual-phase training and introduces the Operational Turing Test benchmark for evaluating operational grounding.

Findings

FMSLT effectively incorporates operational context into tabular data models.

Dual-phase training improves zero-shot inference on proprietary data.

Operational grounding is crucial for autonomous data-driven agents.

Abstract

This position paper argues that foundation models for tabular data face inherent limitations when isolated from operational context - the procedural logic, declarative rules, and domain knowledge that define how data is created and governed. Current approaches focus on single-table generalization or schema-level relationships, fundamentally missing the operational knowledge that gives data meaning. We introduce Semantically Linked Tables (SLT) and Foundation Models for SLT (FMSLT) as a new model class that grounds tabular data in its operational context. We propose dual-phase training: pre-training on open-source code-data pairs and synthetic systems to learn business logic mechanics, followed by zero-shot inference on proprietary data. We introduce the ``Operational Turing Test'' benchmark and argue that operational grounding is essential for autonomous agents in complex data environments.