Cominsight organizes atomic data insights for better report writing
Structured Composition of Verifiable Atomic Insights for Table-to-Report Generation
Computation and Language
Summary
Automatically turning tables of data into detailed reports is tricky because early guesses often limit deeper understanding. The authors propose ComInsight, a method that breaks down data insights into small, verified pieces and connects them logically. This helps create accurate and easy-to-check reports that combine information across tables and perspectives. Their approach also shows exactly which data supports each conclusion, improving trust in the results. Tests show ComInsight makes reports that are more factually correct and complete than existing methods.
What this means in practice
- •For business intelligence teams: Generate clear, verifiable analytical reports from multiple data tables to support decision-making without missing cross-table insights.
- •For financial analysts: Produce detailed, traceable reports combining various financial metrics automatically, improving report reliability for investment decisions.
Authors
Teng Lin, Xinyu Liu, Nan Tang
Abstract
Table-to-report generation refers to the task of automatically generating article-level analyt- ical reports from relational tables and is an essential capability for automated data science and decision support. Its central challenge lies in systematically discovering verifiable com- posite insights across tables, attributes, and analytical perspectives, and organizing them into coherent, complete, and traceable evidence chains. Existing methods primarily rely on sequential, reactive data agents or direct Large Language Model(LLM) generation. They suffer from exploration bias: early local observations constrain subsequent actions, causing models to focus prematurely on local analyzes and miss cross-table or cross-dimensional evidence. We propose ComInsight, which reformulates insight discovery as the composition of atomic evidences. We first define an atomic insight as the smallest executable analytical unit conforming to a predefined analysis pattern and enumerate all valid atomic insights from database schema and content. These atoms are then organized into a multi-relational insight graph, where nodes represent verified data facts and edges encode logical, temporal, or hierarchical relations. Finally, a set of composition operators systematically fuses atomic nodes into higher-order composite conclusions. Every composite output is accompanied by executable SQL and fine-grained provenance, ensuring full verifiability. Across three benchmarks InsightBench, DDR-Bench, and T2R-Bench, ComInsight consistently outperforms strong baselines in factual correctness, novelty, and structural completeness. We believe ComInsight offers a reliable, efficient, and explainable path toward table-to-report generation.