Fetching the paper…

Error Analysis Prompting Enables Human-Like Translation Evaluation in Large Language Models · Around