Natural Language Inference (NLI) or Recognizing Textual Entailment (RTE) aims at predicting the relation between a pair of sentences (premise and hypothesis) as entailment, contradiction or semantic independence. Although deep learning models have shown promising performance for NLI in recent years, they rely on large scale expensive human-annotated datasets. Semi-supervised learning (SSL) is a popular technique for reducing the reliance on human annotation by leveraging unlabeled data for training. However, despite its substantial success on single sentence classification tasks where the challenge in making use of unlabeled data is to assign "good enough" pseudo-labels, for NLI tasks, the nature of unlabeled data is more complex: one of the sentences in the pair (usually the hypothesis) along with the class label are missing from the data and require human annotations, which makes SSL for NLI more challenging. In this paper, we propose a novel way to incorporate unlabeled data in SSL for NLI where we use a conditional language model, BART to generate the hypotheses for the unlabeled sentences (used as premises). Our experiments show that our SSL framework successfully exploits unlabeled data and substantially improves the performance of four NLI datasets in low-resource settings. We release our code at: https://github.com/msadat3/SSL_for_NLI.
The emergence of noisy medium-scale quantum devices has led to proof-of-concept applications for quantum computing in various domains. Examples include Natural Language Processing (NLP) where sentence classification experiments have been carried out, as well as procedural generation, where tasks such as geopolitical map creation, and image manipulation have been performed. We explore applications at the intersection of these two areas by designing a hybrid quantum-classical algorithm for sentence generation. Our algorithm is based on the well-known simulated annealing technique for combinatorial optimisation. An implementation is provided and used to demonstrate successful sentence generation on both simulated and real quantum hardware. A variant of our algorithm can also be used for music generation. This paper aims to be self-contained, introducing all the necessary background on NLP and quantum computing along the way.
Scientific action graphs extraction from materials synthesis procedures is important for reproducible research, machine automation, and material prediction. But the lack of annotated data has hindered progress in this field. We demonstrate an effort to annotate Polycrystalline Materials Synthesis Procedures (PcMSP) from 305 open access scientific articles for the construction of synthesis action graphs. This is a new dataset for material science information extraction that simultaneously contains the synthesis sentences extracted from the experimental paragraphs, as well as the entity mentions and intra-sentence relations. A two-step human annotation and inter-annotator agreement study guarantee the high quality of the PcMSP corpus. We introduce four natural language processing tasks: sentence classification, named entity recognition, relation classification, and joint extraction of entities and relations. Comprehensive experiments validate the effectiveness of several state-of-the-art models for these challenges while leaving large space for improvement. We also perform the error analysis and point out some unique challenges that require further investigation. We will release our annotation scheme, the corpus, and codes to the research community to alleviate the scarcity of labeled data in this domain.
Pre-trained multilingual language models show significant performance gains for zero-shot cross-lingual model transfer on a wide range of natural language understanding (NLU) tasks. Previously, for zero-shot cross-lingual evaluation, pre-trained models are only fine-tuned on English data and tested on a variety of target languages. In this paper, we do cross-lingual evaluation on various NLU tasks (sentence classification, sequence labeling, question answering) using prompt-tuning and compare it with fine-tuning. The results show that prompt tuning achieves much better cross-lingual transfer than fine-tuning across datasets, with only 0.1% to 0.3% tuned parameters. Additionally, we demonstrate through the analysis that prompt tuning can have better cross-lingual transferability of representations on downstream tasks with better aligned decision boundaries.
Many NLP datasets have been found to contain shortcuts: simple decision rules that achieve surprisingly high accuracy. However, it is difficult to discover shortcuts automatically. Prior work on automatic shortcut detection has focused on enumerating features like unigrams or bigrams, which can find only low-level shortcuts, or relied on post-hoc model interpretability methods like saliency maps, which reveal qualitative patterns without a clear statistical interpretation. In this work, we propose to use probabilistic grammars to characterize and discover shortcuts in NLP datasets. Specifically, we use a context-free grammar to model patterns in sentence classification datasets and use a synchronous context-free grammar to model datasets involving sentence pairs. The resulting grammars reveal interesting shortcut features in a number of datasets, including both simple and high-level features, and automatically identify groups of test examples on which conventional classifiers fail. Finally, we show that the features we discover can be used to generate diagnostic contrast examples and incorporated into standard robust optimization methods to improve worst-group accuracy.
Translation has played a crucial role in improving the performance on multilingual tasks: (1) to generate the target language data from the source language data for training and (2) to generate the source language data from the target language data for inference. However, prior works have not considered the use of both translations simultaneously. This paper shows that combining them can synergize the results on various multilingual sentence classification tasks. We empirically find that translation artifacts stylized by translators are the main factor of the performance gain. Based on this analysis, we adopt two training methods, SupCon and MixUp, considering translation artifacts. Furthermore, we propose a cross-lingual fine-tuning algorithm called MUSC, which uses SupCon and MixUp jointly and improves the performance. Our code is available at https://github.com/jongwooko/MUSC.
This paper presents Macquarie University's participation to the two most recent BioASQ Synergy Tasks (as per June 2022), and to the BioASQ10 Task~B (BioASQ10b), Phase~B. In these tasks, participating systems are expected to generate complex answers to biomedical questions, where the answers may contain more than one sentence. We apply query-focused extractive summarisation techniques. In particular, we follow a sentence classification-based approach that scores each candidate sentence associated to a question, and the $n$ highest-scoring sentences are returned as the answer. The Synergy Task corresponds to an end-to-end system that requires document selection, snippet selection, and finding the final answer, but it has very limited training data. For the Synergy task, we selected the candidate sentences following two phases: document retrieval and snippet retrieval, and the final answer was found by using a DistilBERT/ALBERT classifier that had been trained on the training data of BioASQ9b. Document retrieval was achieved as a standard search over the CORD-19 data using the search API provided by the BioASQ organisers, and snippet retrieval was achieved by re-ranking the sentences of the top retrieved documents, using the cosine similarity of the question and candidate sentence. We observed that vectors represented via sBERT have an edge over tf.idf. BioASQ10b Phase B focuses on finding the specific answers to biomedical questions. For this task, we followed a data-centric approach. We hypothesised that the training data of the first BioASQ years might be biased and we experimented with different subsets of the training data. We observed an improvement of results when the system was trained on the second half of the BioASQ10b training data.
A timeline provides one of the most effective ways to visualize the important historical facts that occurred over a period of time, presenting the insights that may not be so apparent from reading the equivalent information in textual form. By leveraging generative adversarial learning for important sentence classification and by assimilating knowledge based tags for improving the performance of event coreference resolution we introduce a two staged system for event timeline generation from multiple (historical) text documents. We demonstrate our results on two manually annotated historical text documents. Our results can be extremely helpful for historians, in advancing research in history and in understanding the socio-political landscape of a country as reflected in the writings of famous personas.
Word Mover's Distance (WMD) computes the distance between words and models text similarity with the moving cost between words in two text sequences. Yet, it does not offer good performance in sentence similarity evaluation since it does not incorporate word importance and fails to take inherent contextual and structural information in a sentence into account. An improved WMD method using the syntactic parse tree, called Syntax-aware Word Mover's Distance (SynWMD), is proposed to address these two shortcomings in this work. First, a weighted graph is built upon the word co-occurrence statistics extracted from the syntactic parse trees of sentences. The importance of each word is inferred from graph connectivities. Second, the local syntactic parsing structure of words is considered in computing the distance between words. To demonstrate the effectiveness of the proposed SynWMD, we conduct experiments on 6 textual semantic similarity (STS) datasets and 4 sentence classification datasets. Experimental results show that SynWMD achieves state-of-the-art performance on STS tasks. It also outperforms other WMD-based methods on sentence classification tasks.
In this paper, we introduce data multiplexing (DataMUX), a technique that enables deep neural networks to process multiple inputs simultaneously using a single compact representation. DataMUX demonstrates that neural networks are capable of generating accurate predictions over mixtures of inputs, resulting in increased throughput with minimal extra memory requirements. Our approach uses two key components -- 1) a multiplexing layer that performs a fixed linear transformation to each input before combining them to create a mixed representation of the same size as a single input, which is then processed by the base network, and 2) a demultiplexing layer that converts the base network's output back into independent representations before producing predictions for each input. We show the viability of DataMUX for different architectures (Transformers, and to a lesser extent MLPs and CNNs) across six different tasks spanning sentence classification, named entity recognition and image classification. For instance, DataMUX for Transformers can multiplex up to $20$x/$40$x inputs, achieving $11$x/$18$x increase in throughput with minimal absolute performance drops of $<2\%$ and $<4\%$ respectively on MNLI, a natural language inference task. We also provide a theoretical construction for multiplexing in self-attention networks and analyze the effect of various design elements in DataMUX.
To improve deep-learning performance in low-resource settings, many researchers have redesigned model architectures or applied additional data (e.g., external resources, unlabeled samples). However, there have been relatively few discussions on how to make good use of small amounts of labeled samples, although it is potentially beneficial and should be done before applying additional data or redesigning models. In this study, we assume a low-resource setting in which only a few labeled samples (i.e., 30-100 per class) are available, and we discuss how to exploit them without additional data or model redesigns. We explore possible approaches in the following three aspects: training-validation splitting, early stopping, and weight initialization. Extensive experiments are conducted on six public sentence classification datasets. Performance on various evaluation metrics (e.g., accuracy, loss, and calibration error) significantly varied depending on the approaches that were combined in the three aspects. Based on the results, we propose an integrated method, which is to initialize the model with a weight averaging method and use a non-validation stop method to train all samples. This simple integrated method consistently outperforms the competitive methods; e.g., the average accuracy of six datasets of this method was 1.8% higher than those of conventional validation-based methods. In addition, the integrated method further improves the performance when adapted to several state-of-the-art models that use additional data or redesign the network architecture (e.g., self-training and enhanced structural models). Our results highlight the importance of the training strategy and suggest that the integrated method can be the first step in the low-resource setting. This study provides empirical knowledge that will be helpful when dealing with low-resource data in future efforts.
We introduce the novel task of detecting sustainability initiatives in company reports. Given a full report, the aim is to automatically identify mentions of practical activities that a company has performed in order to tackle specific societal issues. New methods for identifying continuous sentence spans need to be developed for capturing the multi-sentence structure of individual sustainability initiatives. We release a new dataset of company reports in which the text has been manually annotated with sustainability initiatives. We also evaluate different models for initiative detection, introducing a novel aggregation and evaluation methodology. Our proposed architecture uses sequences of consecutive sentences to account for contextual information when making classification decisions at the individual sentence level.
As unlabeled data carry rich task-relevant information, they are proven useful for few-shot learning of language model. The question is how to effectively make use of such data. In this work, we revisit the self-training technique for language model fine-tuning and present a state-of-the-art prompt-based few-shot learner, SFLM. Given two views of a text sample via weak and strong augmentation techniques, SFLM generates a pseudo label on the weakly augmented version. Then, the model predicts the same pseudo label when fine-tuned with the strongly augmented version. This simple approach is shown to outperform other state-of-the-art supervised and semi-supervised counterparts on six sentence classification and six sentence-pair classification benchmarking tasks. In addition, SFLM only relies on a few in-domain unlabeled data. We conduct a comprehensive analysis to demonstrate the robustness of our proposed approach under various settings, including augmentation techniques, model scale, and few-shot knowledge transfer across tasks.
We propose a novel architecture for Graph Neural Networks that is inspired by the idea behind Tree Kernels of measuring similarity between trees by taking into account their common substructures, named fragments. By imposing a series of regularization constraints to the learning problem, we exploit a pooling mechanism that incorporates such notion of fragments within the node soft assignment function that produces the embeddings. We present an extensive experimental evaluation on a collection of sentence classification tasks conducted on several argument mining corpora, showing that the proposed approach performs well with respect to state-of-the-art techniques.
Domain adaptive pretraining, i.e. the continued unsupervised pretraining of a language model on domain-specific text, improves the modelling of text for downstream tasks within the domain. Numerous real-world applications are based on domain-specific text, e.g. working with financial or biomedical documents, and these applications often need to support multiple languages. However, large-scale domain-specific multilingual pretraining data for such scenarios can be difficult to obtain, due to regulations, legislation, or simply a lack of language- and domain-specific text. One solution is to train a single multilingual model, taking advantage of the data available in as many languages as possible. In this work, we explore the benefits of domain adaptive pretraining with a focus on adapting to multiple languages within a specific domain. We propose different techniques to compose pretraining corpora that enable a language model to both become domain-specific and multilingual. Evaluation on nine domain-specific datasets-for biomedical named entity recognition and financial sentence classification-covering seven different languages show that a single multilingual domain-specific model can outperform the general multilingual model, and performs close to its monolingual counterpart. This finding holds across two different pretraining methods, adapter-based pretraining and full model pretraining.
Deep learning models exhibit a preference for statistical fitting over logical reasoning. Spurious correlations might be memorized when there exists statistical bias in training data, which severely limits the model performance especially in small data scenarios. In this work, we introduce Counterfactual Adversarial Training framework (CAT) to tackle the problem from a causality perspective. Particularly, for a specific sample, CAT first generates a counterfactual representation through latent space interpolation in an adversarial manner, and then performs Counterfactual Risk Minimization (CRM) on each original-counterfactual pair to adjust sample-wise loss weight dynamically, which encourages the model to explore the true causal effect. Extensive experiments demonstrate that CAT achieves substantial performance improvement over SOTA across different downstream tasks, including sentence classification, natural language inference and question answering.
Representation learning for text via pretraining a language model on a large corpus has become a standard starting point for building NLP systems. This approach stands in contrast to autoencoders, also trained on raw text, but with the objective of learning to encode each input as a vector that allows full reconstruction. Autoencoders are attractive because of their latent space structure and generative properties. We therefore explore the construction of a sentence-level autoencoder from a pretrained, frozen transformer language model. We adapt the masked language modeling objective as a generative, denoising one, while only training a sentence bottleneck and a single-layer modified transformer decoder. We demonstrate that the sentence representations discovered by our model achieve better quality than previous methods that extract representations from pretrained transformers on text similarity tasks, style transfer (an example of controlled generation), and single-sentence classification tasks in the GLUE benchmark, while using fewer parameters than large pretrained models.
This paper presents Macquarie University's participation to the BioASQ Synergy Task, and BioASQ9b Phase B. In each of these tasks, our participation focused on the use of query-focused extractive summarisation to obtain the ideal answers to medical questions. The Synergy Task is an end-to-end question answering task on COVID-19 where systems are required to return relevant documents, snippets, and answers to a given question. Given the absence of training data, we used a query-focused summarisation system that was trained with the BioASQ8b training data set and we experimented with methods to retrieve the documents and snippets. Considering the poor quality of the documents and snippets retrieved by our system, we observed reasonably good quality in the answers returned. For phase B of the BioASQ9b task, the relevant documents and snippets were already included in the test data. Our system split the snippets into candidate sentences and used BERT variants under a sentence classification setup. The system used the question and candidate sentence as input and was trained to predict the likelihood of the candidate sentence being part of the ideal answer. The runs obtained either the best or second best ROUGE-F1 results of all participants to all batches of BioASQ9b. This shows that using BERT in a classification setup is a very strong baseline for the identification of ideal answers.
Reading comprehension is a complex cognitive process involving many human brain activities. Plenty of works have studied the patterns and attention allocations of reading comprehension in information retrieval related scenarios. However, little is known about what happens in human brain during reading comprehension and how these cognitive activities can affect information retrieval process. Additionally, with the advances in brain imaging techniques such as electroencephalogram (EEG), it is possible to collect brain signals in almost real time and explore whether it can be utilized as feedback to facilitate information acquisition performance. In this paper, we carefully design a lab-based user study to investigate brain activities during reading comprehension. Our findings show that neural responses vary with different types of reading contents, i.e., contents that can satisfy users' information needs and contents that cannot. We suggest that various cognitive activities, e.g., cognitive loading, semantic-thematic understanding, and inferential processing, underpin these neural responses at the micro-time scale during reading comprehension. From these findings, we illustrate several insights for information retrieval tasks, such as ranking models construction and interface design. Besides, we suggest the possibility of detecting reading comprehension status for a proactive real-world system. To this end, we propose a Unified framework for EEG-based Reading Comprehension Modeling (UERCM). To verify its effectiveness, we conduct extensive experiments based on EEG features for two reading comprehension tasks: answer sentence classification and answer extraction. Results show that it is feasible to improve the performance of two tasks with brain signals.
Active learning is the iterative construction of a classification model through targeted labeling, enabling significant labeling cost savings. As most research on active learning has been carried out before transformer-based language models ("transformers") became popular, despite its practical importance, comparably few papers have investigated how transformers can be combined with active learning to date. This can be attributed to the fact that using state-of-the-art query strategies for transformers induces a prohibitive runtime overhead, which effectively nullifies, or even outweighs the desired cost savings. For this reason, we revisit uncertainty-based query strategies, which had been largely outperformed before, but are particularly suited in the context of fine-tuning transformers. In an extensive evaluation, we connect transformers to experiments from previous research, assessing their performance on five widely used text classification benchmarks. For active learning with transformers, several other uncertainty-based approaches outperform the well-known prediction entropy query strategy, thereby challenging its status as most popular uncertainty baseline in active learning for text classification.