DataMosaic: Explainable and Verifiable Multi-Modal Data Analytics through Extract-Reason-Verify

Zhengxuan Zhang, Zhuowen Liang, Yin Wu, Teng Lin, Yuyu Luo, Nan Tang · 2025

Read DataMosaic: Explainable and Verifiable Multi-Modal Data Analytics through Extract-Reason-Verify by Zhengxuan Zhang, Zhuowen Liang, Yin Wu, Teng Lin, Yuyu Luo, Nan Tang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-10036.txt.

Description

Zhengxuan Zhang, Zhuowen Liang, Yin Wu, Teng Lin, Yuyu Luo, and Nan Tang wrote this paper in two thousand twenty-five. Its text presents DataPuzzle, a plea to rebuild language model data analysis as open steps that can be checked, in place of one-shot prompt answers.

The authors argue that fluency hides a structural flaw: proof, thought, and verdict are crushed into one murky reply. Their frame splits a hard query into parts, casts facts into tables and graphs, and gives set roles to agents for pulling, translating, and linking. Each step can be traced and checked. The piece is a blueprint and a call, not a test report: structure, they say, is the path to trust.

Catalog record

CREATORZhengxuan Zhang, Zhuowen Liang, Yin Wu, Teng Lin, Yuyu Luo, Nan Tang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE27 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-10036

Browse the full library · About the Tower of Babel