ASTRA: Adaptive Semantic Tree Reasoning Architecture for Complex Table Question Answering
Xiaoke Guo, Songze Li, Zhiqiang Liu, Zhaoyan Gong, Yuanxiang Liu + 2 more
TLDR
ASTRA is a new architecture that uses adaptive semantic trees and dual-mode reasoning to significantly improve LLM performance in complex table question answering.
Key contributions
- Introduces ASTRA, an Adaptive Semantic Tree Reasoning Architecture for complex table Q&A.
- Proposes AdaSTR to reconstruct tables into adaptive Logical Semantic Trees, modeling hierarchies.
- Presents DuTR, a dual-mode reasoning framework combining tree-search navigation and code execution.
- Achieves state-of-the-art performance on complex table question answering benchmarks.
Why it matters
This paper tackles a critical limitation of LLMs in complex table question answering: effective table serialization and reasoning. By introducing ASTRA, it provides a novel architecture that explicitly models table hierarchies and offers a robust dual-mode reasoning framework. This significantly advances LLM capabilities for understanding and querying complex tabular data, achieving state-of-the-art results.
Original 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.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.