Enabling Classifiers to Make Judgements Explicitly Aligned with Human Values
Bang, Yejin, Yu, Tiezheng, Madotto, Andrea, Lin, Zhaojiang, Diab, Mona, Fung, Pascale
–arXiv.org Artificial Intelligence
Many NLP classification tasks, such as sexism/racism detection or toxicity detection, are based on human values. Yet, human values can vary under diverse cultural conditions. Therefore, we introduce a framework for value-aligned classification that performs prediction based on explicitly written human values in the command. Along with the task, we propose a practical approach that distills value-aligned knowledge from large-scale language models (LLMs) to construct value-aligned classifiers in two steps. First, we generate value-aligned training data from LLMs by prompt-based few-shot learning. Next, we fine-tune smaller classification models with the generated data for the task. Empirical results show that our VA-Models surpass multiple baselines by at least 15.56% on the F1-score, including few-shot learning with OPT-175B and existing text augmentation methods. We suggest that using classifiers with explicit human value input improves both inclusivity & explainability in AI.
arXiv.org Artificial Intelligence
Oct-14-2022
- Country:
- North America
- Dominican Republic (0.04)
- United States
- Virginia (0.04)
- New York > New York County
- New York City (0.04)
- Europe > Italy
- Calabria > Catanzaro Province > Catanzaro (0.04)
- Asia > China
- Hong Kong (0.04)
- North America
- Genre:
- Research Report
- New Finding (0.66)
- Experimental Study (0.46)
- Research Report
- Industry:
- Health & Medicine (0.93)
- Law (0.88)
- Law Enforcement & Public Safety > Crime Prevention & Enforcement (0.46)
- Technology: