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

arXiv cs.CL / 4/13/2026

📰 NewsSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces ASTRA, an Adaptive Semantic Tree Reasoning Architecture designed to improve Large Language Model performance on complex table question answering by addressing table serialization bottlenecks.
ASTRA’s AdaSTR module rebuilds input tables into Logical Semantic Trees using the LLM’s global semantic awareness, explicitly modeling hierarchical dependencies and adapting its construction strategy to table scale.
ASTRA’s DuTR module combines dual reasoning modes: tree-search-based textual navigation for linguistic alignment and symbolic code execution for verification.
Experiments on complex table benchmarks show ASTRA achieves state-of-the-art performance, indicating better handling of structural information and more reliable reasoning than prior serialization and tree-based methods.

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.