LLMs analyze reviewer scores, finding rebuttals bounded by initial review structure

A new study published on arXiv investigates the impact of author rebuttals on peer-review scores in academic conferences, specifically examining data from ICLR 2024-2025. Researchers utilized LLMs, including Gemini Flash 3.0 and Claude Opus 4.6, to analyze reviewer trajectories and identify patterns in score movement. The findings indicate that while rebuttals can influence scores, the extent of this movement is largely constrained by the initial review structure, with specific exchange features correlating with rebuttal success or failure. AI

IMPACT Provides insights into how LLMs can be used to analyze and understand complex human processes like academic peer review.

RANK_REASON The cluster contains a research paper published on arXiv detailing a study on peer review processes using LLMs. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX Code Finder for Papers
• Claude Opus 4.6
• CORE Recommender
• DagsHub
• Gemini Flash 3.0
• Gotit.pub
• Hugging Face
• Influence Flower
• International Conference on Learning Representations
• ScienceCast

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →