Rationales in the form of manually annotated input spans usually serve as ground truth when evaluating explainability methods in NLP. They are, however, time-consuming and often biased by the annotation process. In this paper, we debate whether human gaze, in the form of webcam-based eye-tracking recordings, poses a valid alternative when evaluating importance scores. We evaluate the additional information provided by gaze data, such as total reading times, gaze entropy, and decoding accuracy with respect to human rationale annotations. We compare WebQAmGaze, a multilingual dataset for information-seeking QA, with attention and explainability-based importance scores for 4 different multilingual Transformer-based language models (mBERT, distil-mBERT, XLMR, and XLMR-L) and 3 languages (English, Spanish, and German). Our pipeline can easily be applied to other tasks and languages. Our findings suggest that gaze data offers valuable linguistic insights that could be leveraged to infer task difficulty and further show a comparable ranking of explainability methods to that of human rationales.

Eye movement features are considered to be direct signals reflecting human attention distribution with a low cost to obtain, inspiring researchers to augment language models with eye-tracking (ET) data. In this study, we select first fixation duration (FFD) and total reading time (TRT) as the cognitive signals to guide Transformer attention in question-answering (QA) tasks. We design three different ET attention masks based on the two features, either collected from human reading events or generated by a gaze-predicting model. We augment BERT and ALBERT models with attention masks structured based on the ET data. We find that augmenting a model with ET data carries linguistic features complementing the information captured by the model. It improves the models’ performance but compromises the stability. Different Transformer models benefit from different types of ET attention masks, while ALBERT performs better than BERT. Moreover, ET data collected from real-life reading events has better model augmenting ability than the model-predicted data.

Eye-tracking-while-reading corpora play a crucial role in the study of human language processing, and, more recently, have been leveraged for cognitively enhancing neural language models. A critical limitation of existing corpora is that they often lack diversity, comprising primarily native speakers. In this study, we expand the eye-tracking-while-reading dataset CopCo, which initially included only Danish L1 readers with and without dyslexia, by incorporating a new dataset of L2 readers with diverse L1 backgrounds. Thus, the extended CopCo corpus constitutes the first eye-tracking-while-reading dataset encompassing neurotypical L1 and L1 readers with dyslexia as well as L2 readers, all reading the same materials. We first provide extensive descriptive statistics of the extended CopCo corpus. Second, we investigate how different degrees of diversity of the training data affect a state-of-the-art generative model of eye movements in reading. Finally, we use this scanpath generation model for gaze-augmented language modeling and investigate the impact of diversity in the training data on the model’s performance on a range of NLP downstream tasks. The code can be found here: https://github.com/norahollenstein/copco-processing.

When humans read a text, their eye movements are influenced by the structural complexity of the input sentences. This cognitive phenomenon holds across languages and recent studies indicate that multilingual language models utilize structural similarities between languages to facilitate cross-lingual transfer. We use sentence-level eye-tracking patterns as a cognitive indicator for structural complexity and show that the multilingual model XLM-RoBERTa can successfully predict varied patterns for 13 typologically diverse languages, despite being fine-tuned only on English data. We quantify the sensitivity of the model to structural complexity and distinguish a range of complexity characteristics. Our results indicate that the model develops a meaningful bias towards sentence length but also integrates cross-lingual differences. We conduct a control experiment with randomized word order and find that the model seems to additionally capture more complex structural information.

The computational cost of transformer-based language models grows quadratically with the sequence length. In this paper, we introduce the selective cache, which stores the selected key-value pairs from the previous context. By selecting important key-value pairs the model makes better use of the cache so that in limited cache size, a longer context history can be stored. We design three kinds of selection methods. The first is based on human language processing. The key-value pairs are selected if they correspond to tokens that are fixated longer, as recorded in eye-tracking-while-reading experiments. We also incorporate the cognitively-inspired selection process into the language model as a trainable process, resulting in two additional methods with improved performance. The selection task is converted into a pruning task so they can be trained with differentiable masks. We demonstrate that the proposed selective cache improves the language modeling performance across different datasets. With the same number of stored key-value pairs (cache size), our selective cache outperforms XL cache and compressive cache by considerable margins.

Integrating human feedback in models can improve the performance of natural language processing (NLP) models. Feedback can be either explicit (e.g. ranking used in training language models) or implicit (e.g. using human cognitive signals in the form of eyetracking). Prior eye tracking and NLP research reveal that cognitive processes, such as human scanpaths, gleaned from human gaze patterns aid in the understanding and performance of NLP models. However, the collection of real eyetracking data for NLP tasks is challenging due to the requirement of expensive and precise equipment coupled with privacy invasion issues. To address this challenge, we propose ScanTextGAN, a novel model for generating human scanpaths over text. We show that ScanTextGAN-generated scanpaths can approximate meaningful cognitive signals in human gaze patterns. We include synthetically generated scanpaths in four popular NLP tasks spanning six different datasets as proof of concept and show that the models augmented with generated scanpaths improve the performance of all downstream NLP tasks.

To gain a better understanding of the linguistic information encoded in character-based language models, we probe the multilingual contextual CANINE model. We design a range of phonetic probing tasks in six Nordic languages, including Faroese as an additional zero-shot instance. We observe that some phonetic information is indeed encoded in the character representations, as consonants and vowels can be well distinguished using a linear classifier. Furthermore, results for the Danish and Norwegian language seem to be worse for the consonant/vowel distinction in comparison to other languages. The information encoded in these representations can also be learned in a zero-shot scenario, as Faroese shows a reasonably good performance in the same vowel/consonant distinction task.

Dyslexia screening in adults is an open challenge since difficulties may not align with standardised tests designed for children. We collect eye-tracking data from natural reading of Danish texts from readers with dyslexia while closely following the experimental design of a corpus of readers without dyslexia. Research suggests that the opaque orthography of the Danish language affects the diagnostic characteristics of dyslexia. To the best of our knowledge, this is the first attempt to classify dyslexia from eye movements during reading in Danish. We experiment with various machine-learning methods, and our best model yields 0.85 F1 score.

We present the second shared task on eye-tracking data prediction of the Cognitive Modeling and Computational Linguistics Workshop (CMCL). Differently from the previous edition, participating teams are asked to predict eye-tracking features from multiple languages, including a surprise language for which there were no available training data. Moreover, the task also included the prediction of standard deviations of feature values in order to account for individual differences between readers.A total of six teams registered to the task. For the first subtask on multilingual prediction, the winning team proposed a regression model based on lexical features, while for the second subtask on cross-lingual prediction, the winning team used a hybrid model based on a multilingual transformer embeddings as well as statistical features.

Character-level information is included in many NLP models, but evaluating the information encoded in character representations is an open issue. We leverage perceptual representations in the form of shape, sound, and color embeddings and perform a representational similarity analysis to evaluate their correlation with textual representations in five languages. This cross-lingual analysis shows that textual character representations correlate strongly with sound representations for languages using an alphabetic script, while shape correlates with featural scripts. We further develop a set of probing classifiers to intrinsically evaluate what phonological information is encoded in character embeddings. Our results suggest that information on features such as voicing are embedded in both LSTM and transformer-based representations.

Relative word importance is a key metric for natural language processing. In this work, we compare human and model relative word importance to investigate if pretrained neural language models focus on the same words as humans cross-lingually. We perform an extensive study using several importance metrics (gradient-based saliency and attention-based) in monolingual and multilingual models, including eye-tracking corpora from four languages (German, Dutch, English, and Russian). We find that gradient-based saliency, first-layer attention, and attention flow correlate strongly with human eye-tracking data across all four languages. We further analyze the role of word length and word frequency in determining relative importance and find that it strongly correlates with length and frequency, however, the mechanisms behind these non-linear relations remain elusive. We obtain a cross-lingual approximation of the similarity between human and computational language processing and insights into the usability of several importance metrics.

Eye movement recordings from reading are one of the richest signals of human language processing. Corpora of eye movements during reading of contextualized running text is a way of making such records available for natural language processing purposes. Such corpora already exist in some languages. We present CopCo, the Copenhagen Corpus of eye tracking recordings from natural reading of Danish texts. It is the first eye tracking corpus of its kind for the Danish language. CopCo includes 1,832 sentences with 34,897 tokens of Danish text extracted from a collection of speech manuscripts. This first release of the corpus contains eye tracking data from 22 participants. It will be extended continuously with more participants and texts from other genres. We assess the data quality of the recorded eye movements and find that the extracted features are in line with related research. The dataset available here: https://osf.io/ud8s5/.

Collecting human judgements is currently the most reliable evaluation method for natural language generation systems. Automatic metrics have reported flaws when applied to measure quality aspects of generated text and have been shown to correlate poorly with human judgements. However, human evaluation is time and cost-intensive, and we lack consensus on designing and conducting human evaluation experiments. Thus there is a need for streamlined approaches for efficient collection of human judgements when evaluating natural language generation systems. Therefore, we present a dynamic approach to measure the required number of human annotations when evaluating generated outputs in relative comparison settings. We propose an agent-based framework of human evaluation to assess multiple labelling strategies and methods to decide the better model in a simulation and a crowdsourcing case study. The main results indicate that a decision about the superior model can be made with high probability across different labelling strategies, where assigning a single random worker per task requires the least overall labelling effort and thus the least cost.

It has been shown that multilingual transformer models are able to predict human reading behavior when fine-tuned on small amounts of eye tracking data. As the cumulated prediction results do not provide insights into the linguistic cues that the model acquires to predict reading behavior, we conduct a deeper analysis of the predictions from the perspective of readability. We try to disentangle the three-fold relationship between human eye movements, the capability of language models to predict these eye movement patterns, and sentence-level readability measures for English. We compare a range of model configurations to multiple baselines. We show that the models exhibit difficulties with function words and that pre-training only provides limited advantages for linguistic generalization.

Human fixation patterns have been shown to correlate strongly with Transformer-based attention. Those correlation analyses are usually carried out without taking into account individual differences between participants and are mostly done on monolingual datasets making it difficult to generalise findings. In this paper, we analyse eye-tracking data from speakers of 13 different languages reading both in their native language (L1) and in English as language learners (L2). We find considerable differences between languages but also that individual reading behaviour such as skipping rate, total reading time and vocabulary knowledge (LexTALE) influence the alignment between humans and models to an extent that should be considered in future studies.

Determining the relative importance of the elements in a sentence is a key factor for effortless natural language understanding. For human language processing, we can approximate patterns of relative importance by measuring reading fixations using eye-tracking technology. In neural language models, gradient-based saliency methods indicate the relative importance of a token for the target objective. In this work, we compare patterns of relative importance in English language processing by humans and models and analyze the underlying linguistic patterns. We find that human processing patterns in English correlate strongly with saliency-based importance in language models and not with attention-based importance. Our results indicate that saliency could be a cognitively more plausible metric for interpreting neural language models. The code is available on github: https://github.com/beinborn/relative_importance.

We analyze if large language models are able to predict patterns of human reading behavior. We compare the performance of language-specific and multilingual pretrained transformer models to predict reading time measures reflecting natural human sentence processing on Dutch, English, German, and Russian texts. This results in accurate models of human reading behavior, which indicates that transformer models implicitly encode relative importance in language in a way that is comparable to human processing mechanisms. We find that BERT and XLM models successfully predict a range of eye tracking features. In a series of experiments, we analyze the cross-domain and cross-language abilities of these models and show how they reflect human sentence processing.

Eye-tracking data from reading represent an important resource for both linguistics and natural language processing. The ability to accurately model gaze features is crucial to advance our understanding of language processing. This paper describes the Shared Task on Eye-Tracking Data Prediction, jointly organized with the eleventh edition of the Work- shop on Cognitive Modeling and Computational Linguistics (CMCL 2021). The goal of the task is to predict 5 different token- level eye-tracking metrics of the Zurich Cognitive Language Processing Corpus (ZuCo). Eye-tracking data were recorded during natural reading of English sentences. In total, we received submissions from 13 registered teams, whose systems include boosting algorithms with handcrafted features, neural models leveraging transformer language models, or hybrid approaches. The winning system used a range of linguistic and psychometric features in a gradient boosting framework.

We demonstrate the functionalities of the new user interface for CogniVal. CogniVal is a framework for the cognitive evaluation of English word embeddings, which evaluates the quality of the embeddings based on their performance to predict human lexical representations from cognitive language processing signals from various sources. In this paper, we present an easy-to-use command line interface for CogniVal with multiple improvements over the original work, including the possibility to evaluate custom embeddings against custom cognitive data sources.

We recorded and preprocessed ZuCo 2.0, a new dataset of simultaneous eye-tracking and electroencephalography during natural reading and during annotation. This corpus contains gaze and brain activity data of 739 English sentences, 349 in a normal reading paradigm and 390 in a task-specific paradigm, in which the 18 participants actively search for a semantic relation type in the given sentences as a linguistic annotation task. This new dataset complements ZuCo 1.0 by providing experiments designed to analyze the differences in cognitive processing between natural reading and annotation. The data is freely available here: https://osf.io/2urht/.

NLP models are imperfect and lack intricate capabilities that humans access automatically when processing speech or reading a text. Human language processing data can be leveraged to increase the performance of models and to pursue explanatory research for a better understanding of the differences between human and machine language processing. We review recent studies leveraging different types of cognitive processing signals, namely eye-tracking, M/EEG and fMRI data recorded during language understanding. We discuss the role of cognitive data for machine learning-based NLP methods and identify fundamental challenges for processing pipelines. Finally, we propose practical strategies for using these types of cognitive signals to enhance NLP models.

We introduce CGA, a conditional VAE architecture, to control, generate, and augment text. CGA is able to generate natural English sentences controlling multiple semantic and syntactic attributes by combining adversarial learning with a context-aware loss and a cyclical word dropout routine. We demonstrate the value of the individual model components in an ablation study. The scalability of our approach is ensured through a single discriminator, independently of the number of attributes. We show high quality, diversity and attribute control in the generated sentences through a series of automatic and human assessments. As the main application of our work, we test the potential of this new NLG model in a data augmentation scenario. In a downstream NLP task, the sentences generated by our CGA model show significant improvements over a strong baseline, and a classification performance often comparable to adding same amount of additional real data.

An interesting method of evaluating word representations is by how much they reflect the semantic representations in the human brain. However, most, if not all, previous works only focus on small datasets and a single modality. In this paper, we present the first multi-modal framework for evaluating English word representations based on cognitive lexical semantics. Six types of word embeddings are evaluated by fitting them to 15 datasets of eye-tracking, EEG and fMRI signals recorded during language processing. To achieve a global score over all evaluation hypotheses, we apply statistical significance testing accounting for the multiple comparisons problem. This framework is easily extensible and available to include other intrinsic and extrinsic evaluation methods. We find strong correlations in the results between cognitive datasets, across recording modalities and to their performance on extrinsic NLP tasks.

Previous research shows that eye-tracking data contains information about the lexical and syntactic properties of text, which can be used to improve natural language processing models. In this work, we leverage eye movement features from three corpora with recorded gaze information to augment a state-of-the-art neural model for named entity recognition (NER) with gaze embeddings. These corpora were manually annotated with named entity labels. Moreover, we show how gaze features, generalized on word type level, eliminate the need for recorded eye-tracking data at test time. The gaze-augmented models for NER using token-level and type-level features outperform the baselines. We present the benefits of eye-tracking features by evaluating the NER models on both individual datasets as well as in cross-domain settings.

Learning attention functions requires large volumes of data, but many NLP tasks simulate human behavior, and in this paper, we show that human attention really does provide a good inductive bias on many attention functions in NLP. Specifically, we use estimated human attention derived from eye-tracking corpora to regularize attention functions in recurrent neural networks. We show substantial improvements across a range of tasks, including sentiment analysis, grammatical error detection, and detection of abusive language.

We present an operational component of a real-world patient triage system. Given a specific patient presentation, the system is able to assess the level of medical urgency and issue the most appropriate recommendation in terms of best point of care and time to treat. We use an attention-based convolutional neural network architecture trained on 600,000 doctor notes in German. We compare two approaches, one that uses the full text of the medical notes and one that uses only a selected list of medical entities extracted from the text. These approaches achieve 79% and 66% precision, respectively, but on a confidence threshold of 0.6, precision increases to 85% and 75%, respectively. In addition, a method to detect warning symptoms is implemented to render the classification task transparent from a medical perspective. The method is based on the learning of attention scores and a method of automatic validation using the same data.

Reliably detecting relevant relations between entities in unstructured text is a valuable resource for knowledge extraction, which is why it has awaken significant interest in the field of Natural Language Processing. In this paper, we present a system for relation classification and extraction based on an ensemble of convolutional and recurrent neural networks that ranked first in 3 out of the 4 Subtasks at SemEval 2018 Task 7. We provide detailed explanations and grounds for the design choices behind the most relevant features and analyze their importance.

Inconsistencies are part of any manually annotated corpus. Automatically finding these inconsistencies and correcting them (even manually) can increase the quality of the data. Past research has focused mainly on detecting inconsistency in syntactic annotation. This work explores new approaches to detecting inconsistency in semantic annotation. Two ranking methods are presented in this paper: a discrepancy ranking and an entropy ranking. Those methods are then tested and evaluated on multiple corpora annotated with multiword expressions and supersense labels. The results show considerable improvements in detecting inconsistency candidates over a random baseline. Possible applications of methods for inconsistency detection are improving the annotation procedure as well as the guidelines and correcting errors in completed annotations.