Abstract:Structured tabular data exist across nearly all fields. Reasoning task over these data aims to answer questions or determine the truthiness of hypothesis sentences by understanding the semantic meaning of a table. While previous works have devoted significant efforts to the tabular reasoning task, they always assume there are sufficient labeled data. However, constructing reasoning samples over tables (and related text) is labor-intensive, especially when the reasoning process is complex. When labeled data is insufficient, the performance of models will suffer an unendurable decline. In this paper, we propose a unified framework for unsupervised complex tabular reasoning (UCTR), which generates sufficient and diverse synthetic data with complex logic for tabular reasoning tasks, assuming no human-annotated data at all. Specifically, we first utilize a random sampling strategy to collect diverse programs of different types and execute them on tables based on a "Program-Executor" module. To bridge the gap between the programs and natural language sentences, we design a powerful "NL-Generator" module to generate natural language sentences with complex logic from these programs. Since a table often occurs with its surrounding texts, we further propose novel "Table-to-Text" and "Text-to-Table" operators to handle joint table-text reasoning scenarios. This way, we can adequately exploit the unlabeled table resources to obtain a well-performed reasoning model under an unsupervised setting. Our experiments cover different tasks (question answering and fact verification) and different domains (general and specific), showing that our unsupervised methods can achieve at most 93% performance compared to supervised models. The impressive performance demonstrates that UCTR can significantly reduce the dependence on manual annotation. Moreover, we also find that it can substantially boost the supervised performance in low-resourced domains as a data augmentation technique.

Incorporating External Knowledge to Enhance Tabular Reasoning

Tabular reasoning via two-stage knowledge injection

Enhancing Tabular Reasoning with Pattern Exploiting Training

Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding

Iteratively Learning Embeddings and Rules for Knowledge Graph Reasoning

Knowledge-Aware Reasoning over Multimodal Semi-structured Tables

Rethinking Tabular Data Understanding with Large Language Models

TabSQLify: Enhancing Reasoning Capabilities of LLMs Through Table Decomposition

GIVE: Structured Reasoning with Knowledge Graph Inspired Veracity Extrapolation

Simple Modifications to Improve Tabular Neural Networks

Unveiling Implicit Table Knowledge with Question-Then-Pinpoint Reasoner for Insightful Table Summarization

Implicit Chain of Thought Reasoning via Knowledge Distillation

Beyond Embeddings: The Promise of Visual Table in Visual Reasoning

Toward a Unified Framework for Unsupervised Complex Tabular Reasoning.

Neural Probabilistic Logic Learning for Knowledge Graph Reasoning

TinyThinker: Distilling Reasoning through Coarse-to-Fine Knowledge Internalization with Self-Reflection

Exploring the Numerical Reasoning Capabilities of Language Models: A Comprehensive Analysis on Tabular Data

Visual Chain of Thought: Bridging Logical Gaps with Multimodal Infillings

TACT: Advancing Complex Aggregative Reasoning with Information Extraction Tools

TabularNet: A Neural Network Architecture for Understanding Semantic Structures of Tabular Data

Explicit Knowledge Incorporation for Visual Reasoning