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

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Guo, Xiaoke, Li, Songze, Liu, Zhiqiang, Gong, Zhaoyan, Liu, Yuanxiang, Chen, Huajun, Zhang, Wen
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918524491923456
author Guo, Xiaoke
Li, Songze
Liu, Zhiqiang
Gong, Zhaoyan
Liu, Yuanxiang
Chen, Huajun
Zhang, Wen
author_facet Guo, Xiaoke
Li, Songze
Liu, Zhiqiang
Gong, Zhaoyan
Liu, Yuanxiang
Chen, Huajun
Zhang, Wen
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2604_08999
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ASTRA: Adaptive Semantic Tree Reasoning Architecture for Complex Table Question Answering
Guo, Xiaoke
Li, Songze
Liu, Zhiqiang
Gong, Zhaoyan
Liu, Yuanxiang
Chen, Huajun
Zhang, Wen
Computation and Language
Artificial Intelligence
Machine Learning
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.
title ASTRA: Adaptive Semantic Tree Reasoning Architecture for Complex Table Question Answering
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2604.08999