Published February 17, 2026
| Version v1.0.0
Software
Open
Multi-Model AI Consensus Pipeline for Automated Data Extraction in Meta-Analysis
Authors/Creators
- 1. Institute of Soil, Water and Environmental Sciences, Agricultural Research Organization – Volcani Center, Israel
Description
A multi-model AI consensus pipeline for automated quantitative data extraction from scientific research papers for meta-analysis. Uses challenge-aware routing and dual-model extraction (Claude Sonnet 4, Kimi K2.5) with Gemini 3 Flash tiebreaker. Validated against three published meta-analysis datasets with formal TOST equivalence testing.
Notes
If you use this software, please cite it as below.
Files
halpernmoshe/Meta-analysis-extractor-agriculture-v1.0.0.zip
Files
(4.1 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:d8a04d2110b92c95e7d8df43118d39c6
|
4.1 MB | Preview Download |
Additional details
Related works
- Is supplement to
- Software: https://github.com/halpernmoshe/Meta-analysis-extractor-agriculture/tree/v1.0.0 (URL)