Researchers find how AI judges decide summary quality step by step

Original: Beyond Scores: Understanding LLM-as-a-Judge Mechanisms in Summarization Evaluation

Source: arxiv.org ↗