![[2601.07611] DIAGPaper: Diagnosing Valid and Specific Weaknesses in Scientific Papers via Multi-Agent Reasoning](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2601.07611] DIAGPaper: Diagnosing Valid and Specific Weaknesses in Scientific Papers via Multi-Agent Reasoning
Summary
DIAGPaper introduces a multi-agent framework for identifying and prioritizing weaknesses in scientific papers, addressing limitations of existing methods by simulating expert review criteria and incorporating author rebuttals.
Why It Matters
This research is significant as it enhances the quality of scientific review processes by providing a more nuanced and validated approach to identifying weaknesses in papers. It addresses biases and improves the prioritization of issues, which can lead to better academic discourse and publication quality.
Key Takeaways
- DIAGPaper uses a multi-agent system to simulate expert review criteria.
- It incorporates author rebuttals to validate identified weaknesses.
- The framework prioritizes weaknesses based on severity, improving user experience.
- Experiments show DIAGPaper outperforms existing methods in identifying specific weaknesses.
- This approach can enhance the quality of scientific reviews and publications.
Computer Science > Artificial Intelligence arXiv:2601.07611 (cs) [Submitted on 12 Jan 2026 (v1), last revised 18 Feb 2026 (this version, v2)] Title:DIAGPaper: Diagnosing Valid and Specific Weaknesses in Scientific Papers via Multi-Agent Reasoning Authors:Zhuoyang Zou, Abolfazl Ansari, Delvin Ce Zhang, Dongwon Lee, Wenpeng Yin View a PDF of the paper titled DIAGPaper: Diagnosing Valid and Specific Weaknesses in Scientific Papers via Multi-Agent Reasoning, by Zhuoyang Zou and 4 other authors View PDF HTML (experimental) Abstract:Paper weakness identification using single-agent or multi-agent LLMs has attracted increasing attention, yet existing approaches exhibit key limitations. Many multi-agent systems simulate human roles at a surface level, missing the underlying criteria that lead experts to assess complementary intellectual aspects of a paper. Moreover, prior methods implicitly assume identified weaknesses are valid, ignoring reviewer bias, misunderstanding, and the critical role of author rebuttals in validating review quality. Finally, most systems output unranked weakness lists, rather than prioritizing the most consequential issues for users. In this work, we propose DIAGPaper, a novel multi-agent framework that addresses these challenges through three tightly integrated modules. The customizer module simulates human-defined review criteria and instantiates multiple reviewer agents with criterion-specific expertise. The rebuttal module introduces author agents that...