Near-Negative Distinction: Giving a Second Life to Human Evaluation Datasets
Laban, Philippe, Wu, Chien-Sheng, Liu, Wenhao, Xiong, Caiming
–arXiv.org Artificial Intelligence
Precisely assessing the progress in natural language generation (NLG) tasks is challenging, and human evaluation to establish a preference in a model's output over another is often necessary. However, human evaluation is usually costly, difficult to reproduce, and non-reusable. In this paper, we propose a new and simple automatic evaluation method for NLG called Near-Negative Distinction (NND) that repurposes prior human annotations into NND tests. In an NND test, an NLG model must place a higher likelihood on a high-quality output candidate than on a near-negative candidate with a known error. Model performance is established by the number of NND tests a model passes, as well as the distribution over task-specific errors the model fails on. Through experiments on three NLG tasks (question generation, question answering, and summarization), we show that NND achieves a higher correlation with human judgments than standard NLG evaluation metrics. We then illustrate NND evaluation in four practical scenarios, for example performing fine-grain model analysis, or studying model training dynamics. Our findings suggest that NND can give a second life to human annotations and provide low-cost NLG evaluation.
arXiv.org Artificial Intelligence
Nov-9-2022
- Country:
- North America > United States
- Texas (0.04)
- California
- San Francisco County > San Francisco (0.04)
- Los Angeles County > Los Angeles (0.04)
- Europe > Belgium
- Brussels-Capital Region > Brussels (0.04)
- Asia > India
- North America > United States
- Genre:
- Research Report > New Finding (1.00)
- Industry:
- Transportation (0.47)
- Information Technology (0.46)
- Technology: