SheetCompass: Hierarchical Relation Graphs for Agentic Spreadsheet Reasoning

Panjing He; Mingyue Cheng; Yucong Luo; Li Li; Xiaohan Zhang

cs.AIAug 14, 2026

SheetCompass: Hierarchical Relation Graphs for Agentic Spreadsheet Reasoning

Authors: Panjing He, Mingyue Cheng, Yucong Luo, Li Li, Xiaohan Zhang

Organizations: State Key Laboratory of Cognitive Intelligence, University of Science and Technology of China Hefei, China

Abstract

Spreadsheets are widely used to organize, analyze, and manipulate semi-structured data, yet automated spreadsheet reasoning remains challenging for large language models (LLMs). Real-world workbooks often contain implicit cross-table associations, fine-grained column dependencies, and complex spatial layouts. Existing methods typically flatten these multidimensional structures into sequential strings, losing important intra-sheet boundaries and inter-sheet semantics. Consequently, LLMs cannot exploit the global spatial context that human experts naturally use when inspecting spreadsheets. We propose SheetCompass, a graph-guided and memory-driven agentic framework for spreadsheet reasoning and automation. SheetCompass explicitly models structural relationships within and across worksheets while maintaining task-relevant information in memory, enabling agents to reason more effectively over complex workbooks.

Figures & tables

Figure 1. A comparison of various methods. Traditional markdown flattening collapses the 2D layout that human visual scanning effortlessly recovers. SheetCompass closes this gap with the hierarchical graph and multi-agent workflows.

Figure 2. An overview of the SheetCompass framework. A flowchart showing the SheetCompass framework, starting from data input to the final alignment output.

Backbone	Method	SCB		SB		SheetRM
Backbone	Method	Exec@1 $\uparrow$	Pass@1 $\uparrow$	Soft $\uparrow$	Hard $\uparrow$	Exec@1 $\uparrow$	Pass@1 $\uparrow$
GPT-4	Binder	/	15.6	1.2	0.0	/	6.3
	VBA	77.8	37.1	8.3	5.2	56.2	2.8
	OS-Copilot	/	50.1	17.0	14.0	63.7	12.3
	SheetCopilot	87.3	44.3	13.8	10.1	68.1	0.0
	SheetAgent	94.1	61.1	17.6	13.5	92.4	31.2

Table 1. Comparison of the performance of SheetCompass with other approaches on the SCB, SB, and SheetRM benchmarks(%).

\uparrow

indicates that a higher value is better. The best results are in bold and the second best are underlined.

Category	Component	SCB		SB		SheetRM
Category	Component	Exec@1 $\uparrow$	Pass@1 $\uparrow$	Soft $\uparrow$	Hard $\uparrow$	Exec@1 $\uparrow$	Pass@1 $\uparrow$
Full Model	SheetCompass	96.4	71.3	24.7	22.0	94.2	52.3
Main Component	w/o Hierarchical Graph	87.2	56.4	18.1	14.5	85.2	41.4
	w/o Multi-agent	89.5	59.1	19.3	15.2	87.4	43.3
	w/o Dual-level Memory	92.6	60.4	21.2	18.8	90.5	44.3
w/ Hierarchical Graph	w/o Structural Edge	90.8	62.3	20.5	16.8	88.7	45.7

Table 2. Ablation study of the SheetCompass framework across different datasets (%). The results quantify the performance contributions of the hierarchical graph, dual-level memory, and multi-agent workflow, coupled with a fine-grained analysis of their internal components.

\uparrow

indicates that higher values represent better performance.

Figure 3. Performance comparison between single-table and multi-table scenarios across various baseline.

Figure 4. Hyperparameter sensitivity analysis for SheetCompass. The bars represent accuracy on SCB (left), SheetRM(middle), SB (right) datasets across different settings for reasoning cycles (

\tau

), confidence threshold (

\alpha

), and seeding coefficient (

\lambda

Figure 5. Distribution of structural and semantic edges across SCB and SB datesets.

Figure 6. A case study of SheetCompass solving a cross-table task. The figure illustrates the transition from spreadsheet to the construction of a hierarchical graph, followed by the iterative reasoning and verification process within the multi-agent.

Spreadsheet Automation

Spreadsheet automation uses AI to understand, create, edit, and validate workbooks while preserving their formulas, structure, and intended use.

See all 20 papers

Towards Robust Real-World Spreadsheet Understanding with Multi-Agent Multi-Format Reasoning
Apr 14, 2026Houxing Ren, Mingjie Zhan, Zimu Lu +4Multi-Agent LLMs LLM Agent Reliability
CUHK MMLab · SenseTime Research · Shenzhen Loop Area Institute +1
Beyond Rows to Reasoning: Agentic Retrieval for Multimodal Spreadsheet Understanding and Editing
Mar 6, 2026Anmol Gulati, Sahil Sen, Waqar Sarguroh +1Agentic RAG Spreadsheet Automation
Commercial Technology and Innovation Office, PricewaterhouseCoopers, U.S.
Sheet as Token: A Graph-Enhanced Representation for Multi-Sheet Spreadsheet Understanding
May 7, 2026Yiming Lei, Yiqi Wang, Yujia Zhang +5Graph-Based Retrieval Information Retrieval
Effyic Technology Co., Ltd · UCLA · McGill University

Apr 14, 2026J/K move · Enter open · S save

Towards Robust Real-World Spreadsheet Understanding with Multi-Agent Multi-Format Reasoning

Houxing Ren, Mingjie Zhan, Zimu Lu, Ke Wang+3

CUHK MMLab · SenseTime Research · Shenzhen Loop Area Institute+1

Multi-Agent LLMs LLM Agent Reliability Multimodal Large Language Models

Open paper AskPDF arXiv

Spreadsheets are central to real-world applications such as enterprise reporting, auditing, and scientific data management. Despite their ubiquity, existing large language model based approaches typically treat tables as plain text, overlooking critical layout cues and visual semantics. Moreover, real-world spreadsheets are often massive in scale, exceeding the input length that LLMs can efficiently process. To address these challenges, we propose SpreadsheetAgent, a two-stage multi-agent framework for spreadsheet understanding that adopts a step-by-step reading and reasoning paradigm. Instead of loading the entire spreadsheet at once, SpreadsheetAgent incrementally interprets localized regions through multiple modalities, including code execution results, images, and LaTeX tables. The method first constructs a structural sketch and row/column summaries, and then performs task-driven reasoning over this intermediate representation in the Solving Stage. To further enhance reliability, we design a verification module that validates extracted structures via targeted inspections, reducing error propagation and ensuring trustworthy inputs for downstream reasoning. Extensive experiments on two spreadsheet datasets demonstrate the effectiveness of our approach. With GPT-OSS-120B, SpreadsheetAgent achieves 38.16% on Spreadsheet Bench, outperforming the ChatGPT Agent baseline (35.27%) by 2.89 absolute points. These results highlight the potential of SpreadsheetAgent to advance robust and scalable spreadsheet understanding in real-world applications. Code is available at https://github.com/renhouxing/SpreadsheetAgent.git.

Abstract

Spreadsheet Automation

Explore similar work

Towards Robust Real-World Spreadsheet Understanding with Multi-Agent Multi-Format Reasoning

Beyond Rows to Reasoning: Agentic Retrieval for Multimodal Spreadsheet Understanding and Editing

Sheet as Token: A Graph-Enhanced Representation for Multi-Sheet Spreadsheet Understanding