Prediction head is a crucial component of Transformer language models. Despite its direct impact on prediction, this component has often been overlooked in analyzing Transformers.In this study, we investigate the inner workings of the prediction head, specifically focusing on bias parameters. Our experiments with BERT and GPT-2 models reveal that the biases in their word prediction heads play a significant role in the models’ ability to reflect word frequency in a corpus, aligning with the logit adjustment method commonly used in long-tailed learning. We also quantify the effect of controlling the biases in practical auto-regressive text generation scenarios;under a particular setting, more diverse text can be generated without compromising text quality.

The performance of sentence encoders can be significantly improved through the simple practice of fine-tuning using contrastive loss. A natural question arises: what characteristics do models acquire during contrastive learning? This paper theoretically and experimentally shows that contrastive-based sentence encoders implicitly weight words based on information-theoretic quantities; that is, more informative words receive greater weight, while others receive less. The theory states that, in the lower bound of the optimal value of the contrastive learning objective, the norm of word embedding reflects the information gain associated with the distribution of surrounding words. We also conduct comprehensive experiments using various models, multiple datasets, two methods to measure the implicit weighting of models (Integrated Gradients and SHAP), and two information-theoretic quantities (information gain and self-information). The results provide empirical evidence that contrastive fine-tuning emphasizes informative words.

Measuring the semantic similarity between two sentences is still an important task. The word mover’s distance (WMD) computes the similarity via the optimal alignment between the sets of word embeddings. However, WMD does not utilize word order, making it challenging to distinguish sentences with significant overlaps of similar words, even if they are semantically very different. Here, we attempt to improve WMD by incorporating the sentence structure represented by BERT’s self-attention matrix (SAM). The proposed method is based on the Fused Gromov-Wasserstein distance, which simultaneously considers the similarity of the word embedding and the SAM for calculating the optimal transport between two sentences. Experiments demonstrate the proposed method enhances WMD and its variants in paraphrase identification with near-equivalent performance in semantic textual similarity.

Distributed representations of words encode lexical semantic information, but what type of information is encoded and how? Focusing on the skip-gram with negative-sampling method, we found that the squared norm of static word embedding encodes the information gain conveyed by the word; the information gain is defined by the Kullback-Leibler divergence of the co-occurrence distribution of the word to the unigram distribution. Our findings are explained by the theoretical framework of the exponential family of probability distributions and confirmed through precise experiments that remove spurious correlations arising from word frequency. This theory also extends to contextualized word embeddings in language models or any neural networks with the softmax output layer. We also demonstrate that both the KL divergence and the squared norm of embedding provide a useful metric of the informativeness of a word in tasks such as keyword extraction, proper-noun discrimination, and hypernym discrimination.

Monolingual word alignment is crucial to model semantic interactions between sentences. In particular, null alignment, a phenomenon in which words have no corresponding counterparts, is pervasive and critical in handling semantically divergent sentences. Identification of null alignment is useful on its own to reason about the semantic similarity of sentences by indicating there exists information inequality. To achieve unbalanced word alignment that values both alignment and null alignment, this study shows that the family of optimal transport (OT), i.e., balanced, partial, and unbalanced OT, are natural and powerful approaches even without tailor-made techniques. Our extensive experiments covering unsupervised and supervised settings indicate that our generic OT-based alignment methods are competitive against the state-of-the-arts specially designed for word alignment, remarkably on challenging datasets with high null alignment frequencies.

Transformer architecture has become ubiquitous in the natural language processing field. To interpret the Transformer-based models, their attention patterns have been extensively analyzed. However, the Transformer architecture is not only composed of the multi-head attention; other components can also contribute to Transformers’ progressive performance. In this study, we extended the scope of the analysis of Transformers from solely the attention patterns to the whole attention block, i.e., multi-head attention, residual connection, and layer normalization. Our analysis of Transformer-based masked language models shows that the token-to-token interaction performed via attention has less impact on the intermediate representations than previously assumed. These results provide new intuitive explanations of existing reports; for example, discarding the learned attention patterns tends not to adversely affect the performance. The codes of our experiments are publicly available.

The problem of estimating the probability distribution of labels has been widely studied as a label distribution learning (LDL) problem, whose applications include age estimation, emotion analysis, and semantic segmentation. We propose a tree-Wasserstein distance regularized LDL algorithm, focusing on hierarchical text classification tasks. We propose predicting the entire label hierarchy using neural networks, where the similarity between predicted and true labels is measured using the tree-Wasserstein distance. Through experiments using synthetic and real-world datasets, we demonstrate that the proposed method successfully considers the structure of labels during training, and it compares favorably with the Sinkhorn algorithm in terms of computation time and memory usage.

Interpretable rationales for model predictions are crucial in practical applications. We develop neural models that possess an interpretable inference process for dependency parsing. Our models adopt instance-based inference, where dependency edges are extracted and labeled by comparing them to edges in a training set. The training edges are explicitly used for the predictions; thus, it is easy to grasp the contribution of each edge to the predictions. Our experiments show that our instance-based models achieve competitive accuracy with standard neural models and have the reasonable plausibility of instance-based explanations.

Understanding the influence of a training instance on a neural network model leads to improving interpretability. However, it is difficult and inefficient to evaluate the influence, which shows how a model’s prediction would be changed if a training instance were not used. In this paper, we propose an efficient method for estimating the influence. Our method is inspired by dropout, which zero-masks a sub-network and prevents the sub-network from learning each training instance. By switching between dropout masks, we can use sub-networks that learned or did not learn each training instance and estimate its influence. Through experiments with BERT and VGGNet on classification datasets, we demonstrate that the proposed method can capture training influences, enhance the interpretability of error predictions, and cleanse the training dataset for improving generalization.

Interpretable rationales for model predictions play a critical role in practical applications. In this study, we develop models possessing interpretable inference process for structured prediction. Specifically, we present a method of instance-based learning that learns similarities between spans. At inference time, each span is assigned a class label based on its similar spans in the training set, where it is easy to understand how much each training instance contributes to the predictions. Through empirical analysis on named entity recognition, we demonstrate that our method enables to build models that have high interpretability without sacrificing performance.

Events in a narrative differ in salience: some are more important to the story than others. Estimating event salience is useful for tasks such as story generation, and as a tool for text analysis in narratology and folkloristics. To compute event salience without any annotations, we adopt Barthes’ definition of event salience and propose several unsupervised methods that require only a pre-trained language model. Evaluating the proposed methods on folktales with event salience annotation, we show that the proposed methods outperform baseline methods and find fine-tuning a language model on narrative texts is a key factor in improving the proposed methods.

Large-scale dialogue datasets have recently become available for training neural dialogue agents. However, these datasets have been reported to contain a non-negligible number of unacceptable utterance pairs. In this paper, we propose a method for scoring the quality of utterance pairs in terms of their connectivity and relatedness. The proposed scoring method is designed based on findings widely shared in the dialogue and linguistics research communities. We demonstrate that it has a relatively good correlation with the human judgment of dialogue quality. Furthermore, the method is applied to filter out potentially unacceptable utterance pairs from a large-scale noisy dialogue corpus to ensure its quality. We experimentally confirm that training data filtered by the proposed method improves the quality of neural dialogue agents in response generation.

One key principle for assessing textual similarity is measuring the degree of semantic overlap between texts by considering the word alignment. Such alignment-based approaches are both intuitive and interpretable; however, they are empirically inferior to the simple cosine similarity between general-purpose sentence vectors. We focus on the fact that the norm of word vectors is a good proxy for word importance, and the angle of them is a good proxy for word similarity. However, alignment-based approaches do not distinguish the norm and direction, whereas sentence-vector approaches automatically use the norm as the word importance. Accordingly, we propose decoupling word vectors into their norm and direction then computing the alignment-based similarity with the help of earth mover’s distance (optimal transport), which we refer to as word rotator’s distance. Furthermore, we demonstrate how to grow the norm and direction of word vectors (vector converter); this is a new systematic approach derived from the sentence-vector estimation methods, which can significantly improve the performance of the proposed method. On several STS benchmarks, the proposed methods outperform not only alignment-based approaches but also strong baselines. The source code is avaliable at https://github.com/eumesy/wrd

Attention is a key component of Transformers, which have recently achieved considerable success in natural language processing. Hence, attention is being extensively studied to investigate various linguistic capabilities of Transformers, focusing on analyzing the parallels between attention weights and specific linguistic phenomena. This paper shows that attention weights alone are only one of the two factors that determine the output of attention and proposes a norm-based analysis that incorporates the second factor, the norm of the transformed input vectors. The findings of our norm-based analyses of BERT and a Transformer-based neural machine translation system include the following: (i) contrary to previous studies, BERT pays poor attention to special tokens, and (ii) reasonable word alignment can be extracted from attention mechanisms of Transformer. These findings provide insights into the inner workings of Transformers.

In this paper, we propose a new kernel-based co-occurrence measure that can be applied to sparse linguistic expressions (e.g., sentences) with a very short learning time, as an alternative to pointwise mutual information (PMI). As well as deriving PMI from mutual information, we derive this new measure from the Hilbert–Schmidt independence criterion (HSIC); thus, we call the new measure the pointwise HSIC (PHSIC). PHSIC can be interpreted as a smoothed variant of PMI that allows various similarity metrics (e.g., sentence embeddings) to be plugged in as kernels. Moreover, PHSIC can be estimated by simple and fast (linear in the size of the data) matrix calculations regardless of whether we use linear or nonlinear kernels. Empirically, in a dialogue response selection task, PHSIC is learned thousands of times faster than an RNN-based PMI while outperforming PMI in accuracy. In addition, we also demonstrate that PHSIC is beneficial as a criterion of a data selection task for machine translation owing to its ability to give high (low) scores to a consistent (inconsistent) pair with other pairs.

This paper presents the first study aimed at capturing stylistic similarity between words in an unsupervised manner. We propose extending the continuous bag of words (CBOW) embedding model (Mikolov et al., 2013b) to learn style-sensitive word vectors using a wider context window under the assumption that the style of all the words in an utterance is consistent. In addition, we introduce a novel task to predict lexical stylistic similarity and to create a benchmark dataset for this task. Our experiment with this dataset supports our assumption and demonstrates that the proposed extensions contribute to the acquisition of style-sensitive word embeddings.