ASTRA: Adaptive Semantic Tree Reasoning Architecture for Complex Table Question Answering

TL;DR

ASTRA introduces an adaptive semantic tree approach for complex table question answering, explicitly modeling hierarchies and combining textual and symbolic reasoning to improve performance.

Contribution

The paper proposes ASTRA, a novel architecture with AdaSTR and DuTR modules, enhancing table serialization and reasoning for LLMs in complex QA tasks.

Findings

Achieves state-of-the-art results on complex table benchmarks.

Explicit hierarchical modeling improves reasoning accuracy.

Combines textual navigation and symbolic verification effectively.

Abstract

Table serialization remains a critical bottleneck for Large Language Models (LLMs) in complex table question answering, hindered by challenges such as structural neglect, representation gaps, and reasoning opacity. Existing serialization methods fail to capture explicit hierarchies and lack schema flexibility, while current tree-based approaches suffer from limited semantic adaptability. To address these limitations, we propose ASTRA (Adaptive Semantic Tree Reasoning Architecture) including two main modules, AdaSTR and DuTR. First, we introduce AdaSTR, which leverages the global semantic awareness of LLMs to reconstruct tables into Logical Semantic Trees. This serialization explicitly models hierarchical dependencies and employs an adaptive mechanism to optimize construction strategies based on table scale. Second, building on this structure, we present DuTR, a dual-mode reasoning framework that integrates tree-search-based textual navigation for linguistic alignment and symbolic code execution for precise verification. Experiments on complex table benchmarks demonstrate that our method achieves state-of-the-art (SOTA) performance.