Hyunjae Lee

Hyunjae Lee, Jaewoong Yun, Hyunjin Choi et al.

Contextualized word embeddings can lead to state-of-the-art performances in natural language understanding. Recently, a pre-trained deep contextualized text encoder such as BERT has shown its potential in improving natural language tasks including abstractive summarization. Existing approaches in dialogue summarization focus on incorporating a large language model into summarization task trained on large-scale corpora consisting of news articles rather than dialogues of multiple speakers. In this paper, we introduce self-supervised methods to compensate shortcomings to train a dialogue summarization model. Our principle is to detect incoherent information flows using pretext dialogue text to enhance BERT's ability to contextualize the dialogue text representations. We build and fine-tune an abstractive dialogue summarization model on a shared encoder-decoder architecture using the enhanced BERT. We empirically evaluate our abstractive dialogue summarizer with the SAMSum corpus, a recently introduced dataset with abstractive dialogue summaries. All of our methods have contributed improvements to abstractive summary measured in ROUGE scores. Through an extensive ablation study, we also present a sensitivity analysis to critical model hyperparameters, probabilities of switching utterances and masking interlocutors.

Hyunjin Choi, Hyunjae Lee, Seongho Joe et al.

Encoded representations from a pretrained deep learning model (e.g., BERT text embeddings, penultimate CNN layer activations of an image) convey a rich set of features beneficial for information retrieval. Embeddings for a particular modality of data occupy a high-dimensional space of its own, but it can be semantically aligned to another by a simple mapping without training a deep neural net. In this paper, we take a simple mapping computed from the least squares and singular value decomposition (SVD) for a solution to the Procrustes problem to serve a means to cross-modal information retrieval. That is, given information in one modality such as text, the mapping helps us locate a semantically equivalent data item in another modality such as image. Using off-the-shelf pretrained deep learning models, we have experimented the aforementioned simple cross-modal mappings in tasks of text-to-image and image-to-text retrieval. Despite simplicity, our mappings perform reasonably well reaching the highest accuracy of 77% on recall@10, which is comparable to those requiring costly neural net training and fine-tuning. We have improved the simple mappings by contrastive learning on the pretrained models. Contrastive learning can be thought as properly biasing the pretrained encoders to enhance the cross-modal mapping quality. We have further improved the performance by multilayer perceptron with gating (gMLP), a simple neural architecture.

Hyunjae Lee

This paper describes Difference-aware Deep continuous prompt for Contrastive Sentence Embeddings (D2CSE) that learns sentence embeddings. Compared to state-of-the-art approaches, D2CSE computes sentence vectors that are exceptional to distinguish a subtle difference in similar sentences by employing a simple neural architecture for continuous prompts. Unlike existing architectures that require multiple pretrained language models (PLMs) to process a pair of the original and corrupted (subtly modified) sentences, D2CSE avoids cumbersome fine-tuning of multiple PLMs by only optimizing continuous prompts by performing multiple tasks -- i.e., contrastive learning and conditional replaced token detection all done in a self-guided manner. D2CSE overloads a single PLM on continuous prompts and greatly saves memory consumption as a result. The number of training parameters in D2CSE is reduced to about 1\% of existing approaches while substantially improving the quality of sentence embeddings. We evaluate D2CSE on seven Semantic Textual Similarity (STS) benchmarks, using three different metrics, namely, Spearman's rank correlation, recall@K for a retrieval task, and the anisotropy of an embedding space measured in alignment and uniformity. Our empirical results suggest that shallow (not too meticulously devised) continuous prompts can be honed effectively for multiple NLP tasks and lead to improvements upon existing state-of-the-art approaches.