A Study of Automatic Metrics for the Evaluation of Natural Language Explanations
| Autor: | Miruna-Adriana Clinciu, Helen Hastie, Arash Eshghi |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Computation and Language Computer science business.industry Computer Science - Artificial Intelligence Natural language generation Bayesian network Robotics computer.software_genre Field (computer science) Artificial Intelligence (cs.AI) Transparency (graphic) Key (cryptography) Artificial intelligence business computer Parallels Computation and Language (cs.CL) Natural language processing Natural language |
| Zdroj: | EACL |
| DOI: | 10.48550/arxiv.2103.08545 |
| Popis: | As transparency becomes key for robotics and AI, it will be necessary to evaluate the methods through which transparency is provided, including automatically generated natural language (NL) explanations. Here, we explore parallels between the generation of such explanations and the much-studied field of evaluation of Natural Language Generation (NLG). Specifically, we investigate which of the NLG evaluation measures map well to explanations. We present the ExBAN corpus: a crowd-sourced corpus of NL explanations for Bayesian Networks. We run correlations comparing human subjective ratings with NLG automatic measures. We find that embedding-based automatic NLG evaluation methods, such as BERTScore and BLEURT, have a higher correlation with human ratings, compared to word-overlap metrics, such as BLEU and ROUGE. This work has implications for Explainable AI and transparent robotic and autonomous systems. Comment: Accepted at EACL 2021 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|