We describe an approach for unsupervised learning of a generic, distributed sentence encoder. Using the continuity of text from books, we train an encoder-decoder model that tries to reconstruct the surrounding sentences of an encoded passage. Sentences that share semantic and syntactic properties are thus mapped to similar vector representations. We next introduce a simple vocabulary expansion method to encode words that were not seen as part of training, allowing us to expand our vocabulary to a million words. After training our model, we extract and evaluate our vectors with linear models on 8 tasks: semantic relatedness, paraphrase detection, image-sentence ranking, question-type classification and 4 benchmark sentiment and subjectivity datasets. The end result is an off-the-shelf encoder that can produce highly generic sentence representations that are robust and perform well in practice. We will make our encoder publicly available.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Language Modelling | SICK | MSE | 0.2687 | combine-skip (Kiros et al., 2015) |
| Language Modelling | SICK | Pearson Correlation | 0.8584 | combine-skip (Kiros et al., 2015) |
| Language Modelling | SICK | Spearman Correlation | 0.7916 | combine-skip (Kiros et al., 2015) |
| Sentence Pair Modeling | SICK | MSE | 0.2687 | combine-skip (Kiros et al., 2015) |
| Sentence Pair Modeling | SICK | Pearson Correlation | 0.8584 | combine-skip (Kiros et al., 2015) |
| Sentence Pair Modeling | SICK | Spearman Correlation | 0.7916 | combine-skip (Kiros et al., 2015) |
| Semantic Similarity | SICK | MSE | 0.2687 | combine-skip (Kiros et al., 2015) |
| Semantic Similarity | SICK | Pearson Correlation | 0.8584 | combine-skip (Kiros et al., 2015) |
| Semantic Similarity | SICK | Spearman Correlation | 0.7916 | combine-skip (Kiros et al., 2015) |