Boosting Few-Shot Text Classification via Distribution Estimation
This work addresses few-shot text classification, a key problem in NLP for scenarios with limited labeled data, by adapting a computer vision technique with novel strategies to handle domain differences.
The paper tackles the challenge of applying distribution estimation to few-shot text classification by proposing strategies that use unlabeled query samples to estimate novel class distributions, avoiding negative transfer. The method significantly outperforms state-of-the-art baselines across eight datasets.
Distribution estimation has been demonstrated as one of the most effective approaches in dealing with few-shot image classification, as the low-level patterns and underlying representations can be easily transferred across different tasks in computer vision domain. However, directly applying this approach to few-shot text classification is challenging, since leveraging the statistics of known classes with sufficient samples to calibrate the distributions of novel classes may cause negative effects due to serious category difference in text domain. To alleviate this issue, we propose two simple yet effective strategies to estimate the distributions of the novel classes by utilizing unlabeled query samples, thus avoiding the potential negative transfer issue. Specifically, we first assume a class or sample follows the Gaussian distribution, and use the original support set and the nearest few query samples to estimate the corresponding mean and covariance. Then, we augment the labeled samples by sampling from the estimated distribution, which can provide sufficient supervision for training the classification model. Extensive experiments on eight few-shot text classification datasets show that the proposed method outperforms state-of-the-art baselines significantly.