Learning to Generate Reviews and Discovering Sentiment
This work addresses the challenge of interpretability and data efficiency in natural language processing for researchers and practitioners, though it is incremental as it builds on existing language model techniques.
The paper tackled the problem of discovering high-level concepts like sentiment in unsupervised byte-level recurrent language models, achieving state-of-the-art results on the binary Stanford Sentiment Treebank with data-efficient performance using only a few labeled examples.
We explore the properties of byte-level recurrent language models. When given sufficient amounts of capacity, training data, and compute time, the representations learned by these models include disentangled features corresponding to high-level concepts. Specifically, we find a single unit which performs sentiment analysis. These representations, learned in an unsupervised manner, achieve state of the art on the binary subset of the Stanford Sentiment Treebank. They are also very data efficient. When using only a handful of labeled examples, our approach matches the performance of strong baselines trained on full datasets. We also demonstrate the sentiment unit has a direct influence on the generative process of the model. Simply fixing its value to be positive or negative generates samples with the corresponding positive or negative sentiment.