2021/01/05 by Lorenzo De Mattei, De Mattei, Lorenzo, Michele Cafagna +10
Computer Science · Engineering · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Human Motion and Animation #Multimodal Machine Learning Applications #Video Analysis and Summarization #cs.CL
paper · pdf · doi:10.48550/arxiv.2101.01634
arxiv created 2021/01/05 · openalex publication_date 2021/01/05 · arxiv updated 2021/01/06 · openalex created_date 2021/01/18 · openalex updated_date 2026/07/29
An ongoing debate in the NLG community concerns the best way to evaluate systems, with human evaluation often being considered the most reliable method, compared to corpus-based metrics. However, tasks involving subtle textual differences, such as style transfer, tend to be hard for humans to perform. In this paper, we propose an evaluation method for this task based on purposely-trained classifiers, showing that it better reflects system differences than traditional metrics such as BLEU and ROUGE.