Co-occurring keywords · papers mentioning both
Papers
Breaking the Reviewer: Assessing the Vulnerability of Large Language Models in Automated Peer Review Under Textual Adversarial Attacks
EMNLP 2025
DeepReview: Improving LLM-based Paper Review with Human-like Deep Thinking Process
ACL 2025
Automated Peer Reviewing in Paper SEA: Standardization, Evaluation, and Analysis
EMNLP 2024