Graphy'our Data: Towards End-to-End Modeling, Exploring and Generating Report from Raw Data
Longbin Lai, Changwei Luo, Yunkai Lou, Mingchen Ju, Zhengyi Yang
ACM SIGMOD International Conference on Management of Data Companion (SIGMOD Companion)
RAIDS Lab Authors
Details
Research Area
Tags
Resources
Abstract
While Large Language Models (LLMs) excel at single-document queries and conversational workflows, they struggle with progressively exploring, analyzing, and synthesizing large unstructured document sets, such as in literature surveys. We address this challenge -- termed Progressive Document Investigation -- by introducing Graphy, an end-to-end platform that automates data modeling, exploration and high-quality report generation in a user-friendly manner. Graphy comprises an offline Scrapper that transforms raw documents into a graph, and an online Surveyor that enables iterative exploration and LLM-driven report generation. We showcase a pre-scrapped graph of over 50,000 papers, demonstrating how Graphy facilitates the literature-survey scenario, with video available at https://youtu.be/uM4nzkAdGlM.
Author Affiliations
BibTeX
@inproceedings{lai2025graphy,
title = {Graphy'our Data: Towards End-to-End Modeling, Exploring and Generating Report from Raw Data},
author = {Lai, Longbin and Luo, Changwei and Lou, Yunkai and Ju, Mingchen and Yang, Zhengyi},
series = {SIGMOD/PODS '25},
url = {http://dx.doi.org/10.1145/3722212.3725106},
doi = {10.1145/3722212.3725106},
booktitle = {Companion of the 2025 International Conference on Management of Data},
publisher = {ACM},
year = {2025},
month = June,
pages = {147-150},
collection = {SIGMOD/PODS '25}
}