NoPPA: Non-Parametric Pairwise Attention Random Walk Model for Sentence Representation
This work addresses the need for efficient, trainable-free sentence representations in NLP, offering an incremental improvement over existing non-parametric methods.
The authors tackled the problem of generating sentence embeddings without training by proposing NoPPA, a non-parametric model using pre-trained word embeddings and frequencies, which outperformed bag-of-words methods on eight classification tasks and achieved competitive results with state-of-the-art non-parametric methods.
We propose a novel non-parametric/un-trainable language model, named Non-Parametric Pairwise Attention Random Walk Model (NoPPA), to generate sentence embedding only with pre-trained word embedding and pre-counted word frequency. To the best we know, this study is the first successful attempt to break the constraint on bag-of-words assumption with a non-parametric attention mechanism. We evaluate our method on eight different downstream classification tasks. The experiment results show that NoPPA outperforms all kinds of bag-of-words-based methods in each dataset and provides a comparable or better performance than the state-of-the-art non-parametric methods on average. Furthermore, visualization supports that NoPPA can understand contextual topics, common phrases, and word causalities. Our model is available at https://github.com/JacksonWuxs/NoPPA.