Research manuscript review detection method and system based on error discovery
By constructing an academic paper review model based on structured sample papers and preset evaluation prompts, the problem of error detection in scientific research review has been solved, achieving efficient and accurate paper error detection and evaluation, and improving the accuracy and standardization of scientific research review.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- INST OF AUTOMATION CHINESE ACAD OF SCI
- Filing Date
- 2026-01-30
- Publication Date
- 2026-05-29
AI Technical Summary
Existing technologies lack specific designs for error detection in scientific research peer review, making it difficult to effectively identify logical loopholes, data errors, and methodological defects. They are also susceptible to external factors and lack the ability for multi-round iterative verification and precise location.
We construct an academic paper peer review model based on structured sample papers and preset peer review evaluation prompts. Through supervised fine-tuning and reinforcement learning, we train a large language model to identify various types of errors in scientific papers, including evidence data integrity and methodological logic consistency. We then use a multi-modal error detection framework for accurate peer review.
It has improved the ability to identify errors in scientific research papers, provided standardized defect data support for scientific research peer review standards, and achieved efficient and accurate paper error detection and evaluation.
Smart Images
Figure CN122113907A_ABST