Sequence Classification with Human Attention

Barrett, Maria; Bingel, Joachim; Hollenstein, Nora; Rei, Marek; Søgaard, Anders

doi:10.18653/v1/k18-1030

Cited by 93 publications

(100 citation statements)

References 48 publications

Supporting

Mentioning

100

Contrasting

Order By: Relevance

“…This also opens the question of whether a different architecture could better suit the purpose of leveraging the gaze information in a consistent way. In this context, a potential line of work could adapt human-attention approaches (Barrett et al, 2018) for structured prediction and word-level classification, although it would come at a cost of speed for parsing as sequence labeling (Strzyz et al, 2019b).…”

Section: Discussionmentioning

confidence: 99%

“…Our work focuses on exploiting the utility of gaze information using just a standard BILSTM, directly building on top of previous work of dependency parsing as sequence labeling (Strzyz et al, 2019b), and ignoring extra tools such as attention. In this line, a future possible solution could be to apply the approach by Barrett et al (2018) to structured prediction and word-level classification. In their work they used human data as an inductive bias to update the attention weights of the network.…”

Section: Gaze Informationmentioning

confidence: 99%

“…More in the line of our work, Barrett et al (2018) is one of the few approaches that does not rely on the assumption of using gaze features as input. Instead, the human data is used as an inductive bias to guide the attention weights in a recurrent neural network for sequence classification (used for tasks such as binary sentiment anal-ysis a the sentence level).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Towards Making a Dependency Parser See

Strzyz¹,

Vilares²,

Gómez-Rodríguez³

2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

We explore whether it is possible to leverage eye-tracking data in an RNN dependency parser (for English) when such information is only available during training -i.e. no aggregated or token-level gaze features are used at inference time. To do so, we train a multitask learning model that parses sentences as sequence labeling and leverages gaze features as auxiliary tasks. Our method also learns to train from disjoint datasets, i.e. it can be used to test whether already collected gaze features are useful to improve the performance on new non-gazed annotated treebanks. Accuracy gains are modest but positive, showing the feasibility of the approach. It can serve as a first step towards architectures that can better leverage eye-tracking data or other complementary information available only for training sentences, possibly leading to improvements in syntactic parsing.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Gaze Informationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Towards Making a Dependency Parser See

Strzyz¹,

Vilares²,

Gómez-Rodríguez³

2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

show abstract

“…In this paper, we merely use the TRT feature, which represents total human attention on words during reading. This feature is also used by Carpenter and Just (1983) and Barrett et al (2018). We then di-vide TRT values by the number of participants to get an average TRT (ATRT).…”

Section: Eye-tracking Corpusmentioning

confidence: 99%

“…The reading time of per-word is the indicative of textual (as well as lexical, syntactic and semantic) processing (Demberg and Keller, 2008), which reflects human attention on various content. To obtain human attention during reading, this paper estimates eye fixation duration from eye-tracking corpus inspired by Carpenter and Just (1983) and Barrett et al (2018). The modern-day eye tracking equipment resulting in a very rich and detailed dataset (Cop et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

Using Human Attention to Extract Keyphrase from Microblog Post

Zhang¹,

Zhang²

2019

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

View full text Add to dashboard Cite

This paper studies automatic keyphrase extraction on social media. Previous works have achieved promising results on it, but they neglect human reading behavior during keyphrase annotating. The human attention is a crucial element of human reading behavior. It reveals the relevance of words to the main topics of the target text. Thus, this paper aims to integrate human attention into keyphrase extraction models. First, human attention is represented by the reading duration estimated from eye-tracking corpus. Then, we merge human attention with neural network models by an attention mechanism. In addition, we also integrate human attention into unsupervised models. To the best of our knowledge, we are the first to utilize human attention on keyphrase extraction tasks. The experimental results show that our models have significant improvements on two Twitter datasets.

show abstract

Enhancing keyphrase extraction from microblogs using human reading time

Zhang

2020

Asso for Info Science & Tech

View full text Add to dashboard Cite

The premise of manual keyphrase annotation is to read the corresponding content of an annotated object. Intuitively, when we read, more important words will occupy a longer reading time. Hence, by leveraging human reading time, we can find the salient words in the corresponding content. However, previous studies on keyphrase extraction ignore human reading features. In this article, we aim to leverage human reading time to extract keyphrases from microblog posts. There are two main tasks in this study. One is to determine how to measure the time spent by a human on reading a word. We use eye fixation durations (FDs) extracted from an open source eye‐tracking corpus. Moreover, we propose strategies to make eye FD more effective on keyphrase extraction. The other task is to determine how to integrate human reading time into keyphrase extraction models. We propose two novel neural network models. The first is a model in which the human reading time is used as the ground truth of the attention mechanism. In the second model, we use human reading time as the external feature. Quantitative and qualitative experiments show that our proposed models yield better performance than the baseline models on two microblog datasets.

show abstract

Sequence Classification with Human Attention

Cited by 93 publications

References 48 publications

Towards Making a Dependency Parser See

Towards Making a Dependency Parser See

Using Human Attention to Extract Keyphrase from Microblog Post

Enhancing keyphrase extraction from microblogs using human reading time

Contact Info

Product

Resources

About