Researchers have developed a deep learning approach to analyze sentiment evolution in multi-round peer reviews, a topic previously underexplored. By segmenting review comments from 11,063 papers in Nature Communications and training models on a manually annotated corpus, they identified trends in aspect-level sentiments. The LCF-BERT-CDM model achieved a Macro-F1 score of 82.65%. Findings indicate that as review rounds increase, positive sentiment rises while negative sentiment declines, with aspects like "experiments" and "research significance" showing stronger correlations with the number of rounds. AI
IMPACT Provides a novel method for analyzing scientific discourse, potentially improving review processes and understanding research trends.
RANK_REASON The cluster contains an academic paper detailing a new deep learning approach and its findings.
Read on arXiv cs.IR (Information Retrieval) →
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
