TabularGSM: Understanding the Limitations of LLMs in Tabular Math Reasoning

• URL: http://arxiv.org/abs/2505.19563v2
• Date: Sat, 27 Sep 2025 08:13:03 GMT
• Title: TabularGSM: Understanding the Limitations of LLMs in Tabular Math Reasoning
• Authors: Shi-Yu Tian, Zhi Zhou, Wei Dong, Kun-Yang Yu, Ming Yang, Zi-Jian Cheng, Lan-Zhe Guo, Yu-Feng Li,
• Abstract summary: We propose AutoT2T, a neuro-symbolic framework that transforms math word problems into scalable and verified tabular reasoning tasks.<n>We develop Tabular, a benchmark comprising three progressively complex subsets and a trap subset, with two complementary evaluation settings.
• Score: 26.230588166759706
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Mathematical reasoning has long been a key benchmark for evaluating large language models (LLMs). Although substantial progress has been made on math word problems, the need for reasoning over tabular data in real-world applications has been overlooked. For instance, applications such as business intelligence demand not only multi-step numerical reasoning with tables but also robustness to incomplete or inconsistent information. However, comprehensive evaluation in this area is severely limited, constrained by the reliance on manually collected tables that are difficult to scale and the lack of coverage for potential traps encountered in real-world scenarios. To address this problem, we propose AutoT2T, a neuro-symbolic framework that controllably transforms math word problems into scalable and verified tabular reasoning tasks, enabling the evaluation of both accuracy and robustness. Building on this pipeline, we develop TabularGSM, a benchmark comprising three progressively complex subsets and a trap subset, with two complementary evaluation settings. Our study reveals three key observations: (1) Tabular structure makes mathematical reasoning more challenging; (2) The difficulties stem from the joint effects of tabular retrieval and reasoning; (3) Reasoning robustness is another significant issue that needs to be addressed in existing LLMs. In-depth analyses are conducted for each observation to guide future research.

Related papers

• From Abstract to Contextual: What LLMs Still Cannot Do in Mathematics [79.81905350372067]
  We study gap through contextual mathematical reasoning.<n>We introduce ContextMATH, a benchmark that repurposes AIME and MATH-500 problems into two contextual settings.<n>Open-source models decline by 13 and 34 points on SG and CS, while proprietary models drop by 13 and 20.
  arXiv  Detail & Related papers  (2026-01-30T14:56:04Z)
• GRADE: Generating multi-hop QA and fine-gRAined Difficulty matrix for RAG Evaluation [5.002953635224383]
  Retrieval-Augmented Generation (RAG) systems are widely adopted in knowledge-intensive NLP tasks.<n>Current evaluations often overlook the structural complexity and multi-step reasoning required in real-world scenarios.<n>We propose textscGRADE, a novel evaluation framework that models task difficulty along two dimensions.
  arXiv  Detail & Related papers  (2025-08-23T11:26:41Z)
• LLM-Symbolic Integration for Robust Temporal Tabular Reasoning [69.27153114778748]
  We introduce TempTabQA-C, a synthetic dataset designed for systematic and controlled evaluations.<n>This structured approach allows Large Language Models (LLMs) to generate and executesql queries, enhancing generalization and mitigating biases.
  arXiv  Detail & Related papers  (2025-06-06T05:14:04Z)
• Challenging the Boundaries of Reasoning: An Olympiad-Level Math Benchmark for Large Language Models [86.45058529521258]
  OlymMATH is a novel Olympiad-level mathematical benchmark designed to rigorously test the complex reasoning capabilities of LLMs.<n>OlymMATH features 200 meticulously curated problems, each manually verified and available in parallel English and Chinese versions.
  arXiv  Detail & Related papers  (2025-03-27T11:20:17Z)
• Path-of-Thoughts: Extracting and Following Paths for Robust Relational Reasoning with Large Language Models [62.12031550252253]
  We present Path-of-Thoughts (PoT), a novel framework designed to tackle relation reasoning.<n>PoT efficiently extracts a task-agnostic graph that identifies crucial entities, relations, and attributes within the problem context.<n>PoT identifies relevant reasoning chains within the graph corresponding to the posed question, facilitating inference of potential answers.
  arXiv  Detail & Related papers  (2024-12-23T20:27:12Z)
• TQA-Bench: Evaluating LLMs for Multi-Table Question Answering with Scalable Context and Symbolic Extension [8.489816179329832]
  We present TQA-Bench, a new multi-table QA benchmark designed to evaluate the capabilities of large language models (LLMs) in tackling complex QA tasks over relational data.<n>Our benchmark incorporates diverse relational database instances sourced from real-world public datasets.<n>We systematically evaluate a range of LLMs, both open-source and closed-source, spanning model scales from 7 billion to 70 billion parameters.
  arXiv  Detail & Related papers  (2024-11-29T06:48:13Z)
• TableRAG: Million-Token Table Understanding with Language Models [53.039560091592215]
  TableRAG is a Retrieval-Augmented Generation (RAG) framework specifically designed for LM-based table understanding.<n>TableRAG leverages query expansion combined with schema and cell retrieval to pinpoint crucial information before providing it to the LMs.<n>Our results demonstrate that TableRAG achieves the highest retrieval quality, leading to the new state-of-the-art performance on large-scale table understanding.
  arXiv  Detail & Related papers  (2024-10-07T04:15:02Z)
• Seek and Solve Reasoning for Table Question Answering [49.006950918895306]
  This paper reveals that the reasoning process during task simplification may be more valuable than the simplified tasks themselves.<n>We propose a Seek-and-solving pipeline that instructs the LLM to first seek relevant information and then answer questions.<n>We distill a single-step TQA-solving prompt from this pipeline, using demonstrations with SS-CoT paths to guide the LLM in solving complex TQA tasks.
  arXiv  Detail & Related papers  (2024-09-09T02:41:00Z)
• H-STAR: LLM-driven Hybrid SQL-Text Adaptive Reasoning on Tables [56.73919743039263]
  This paper introduces a novel algorithm that integrates both symbolic and semantic (textual) approaches in a two-stage process to address limitations.<n>Our experiments demonstrate that H-STAR significantly outperforms state-of-the-art methods across three question-answering (QA) and fact-verification datasets.
  arXiv  Detail & Related papers  (2024-06-29T21:24:19Z)
• Exploring the Robustness of Language Models for Tabular Question Answering via Attention Analysis [7.486549276995143]
  Large Language Models (LLMs) have been shown to tackle table comprehension tasks without specific training.<n>We probe how in-context learning (ICL), model scale, instruction tuning, and domain bias affect Tabular QA (TQA)<n>We reveal a strong correlation between perturbation-induced shifts in attention dispersion and the drops in performance.
  arXiv  Detail & Related papers  (2024-06-18T15:41:15Z)
• TAP4LLM: Table Provider on Sampling, Augmenting, and Packing Semi-structured Data for Large Language Model Reasoning [55.33939289989238]
  We propose TAP4LLM as a versatile pre-processor suite for leveraging large language models (LLMs) in table-based tasks effectively. It covers several distinct components: (1) table sampling to decompose large tables into manageable sub-tables based on query semantics, (2) table augmentation to enhance tables with additional knowledge from external sources or models, and (3) table packing & serialization to convert tables into various formats suitable for LLMs' understanding.
  arXiv  Detail & Related papers  (2023-12-14T15:37:04Z)
• MuSR: Testing the Limits of Chain-of-thought with Multistep Soft Reasoning [63.80739044622555]
  We introduce MuSR, a dataset for evaluating language models on soft reasoning tasks specified in a natural language narrative. This dataset has two crucial features. First, it is created through a novel neurosymbolic synthetic-to-natural generation algorithm. Second, our dataset instances are free text narratives corresponding to real-world domains of reasoning.
  arXiv  Detail & Related papers  (2023-10-24T17:59:20Z)
• Large Language Models are Versatile Decomposers: Decompose Evidence and Questions for Table-based Reasoning [45.013230888670435]
  We exploit large language models (LLMs) as decomposers for effective table-based reasoning. We decompose huge evidence (a huge table) into sub-evidence (a small table) to mitigate the interference of useless information. We propose a "parsing-execution-filling" strategy to alleviate the dilemma of the chain of thought.
  arXiv  Detail & Related papers  (2023-01-31T17:51:45Z)
• Optimization Techniques for Unsupervised Complex Table Reasoning via Self-Training Framework [5.351873055148804]
  Self-training framework generates diverse synthetic data with complex logic. We optimize the procedure using a "Table-Text Manipulator" to handle joint table-text reasoning scenarios. UCTRST achieves above 90% of the supervised model performance on different tasks and domains.
  arXiv  Detail & Related papers  (2022-12-20T09:15:03Z)
• Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning [150.17907456113537]
  We present Tabular Math Word Problems (TabMWP), a new dataset containing 38,431 grade-level problems that require mathematical reasoning. We evaluate different pre-trained models on TabMWP, including the GPT-3 model in a few-shot setting. We propose a novel approach, PromptPG, which utilizes policy gradient to learn to select in-context examples from a small amount of training data.
  arXiv  Detail & Related papers  (2022-09-29T08:01:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.