A Multi-Agent LLM Framework for Rating the Quality of Surgical Feedback

Researchers have introduced two new LLM-based frameworks to improve the assessment of surgical feedback quality and phishing email detection.

A new framework uses multi-agent prompting and surgical domain knowledge to assess the quality of verbal feedback from attending surgeons, crucial for resident trainee skill acquisition^[1]. This framework outperformed previous content-based approaches in predicting feedback effectiveness, according to an evaluation of 4.2k trainer feedback instances^[1]. Meanwhile, a separate study presented MultiPhishGuard, a multi-agent LLM system for phishing email detection, achieving an accuracy of 97.89% with a false positive rate of 2.73% and a false negative rate of 0.20%^[2]. Traditional phishing detection methods struggle to keep pace with evolving tactics, but MultiPhishGuard's multi-agent framework with learned coordination and adversarial training improved robustness. Existing methods for assessing surgical feedback quality are either manual or automated but lack nuance, the first study found. The proposed framework discovers interpretable feedback quality criteria grounded in surgical training context.