Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

"Sentiment": models, code, and papers

Enhancing Content Preservation in Text Style Transfer Using Reverse Attention and Conditional Layer Normalization

Aug 01, 2021
Dongkyu Lee, Zhiliang Tian, Lanqing Xue, Nevin L. Zhang

Text style transfer aims to alter the style (e.g., sentiment) of a sentence while preserving its content. A common approach is to map a given sentence to content representation that is free of style, and the content representation is fed to a decoder with a target style. Previous methods in filtering style completely remove tokens with style at the token level, which incurs the loss of content information. In this paper, we propose to enhance content preservation by implicitly removing the style information of each token with reverse attention, and thereby retain the content. Furthermore, we fuse content information when building the target style representation, making it dynamic with respect to the content. Our method creates not only style-independent content representation, but also content-dependent style representation in transferring style. Empirical results show that our method outperforms the state-of-the-art baselines by a large margin in terms of content preservation. In addition, it is also competitive in terms of style transfer accuracy and fluency.

* Accepted to ACL-IJCNLP 2021 

  Access Paper or Ask Questions

Toward Automated Generation of Affective Gestures from Text:A Theory-Driven Approach

Mar 04, 2021
Micol Spitale, Maja J Matarić

Communication in both human-human and human-robot interac-tion (HRI) contexts consists of verbal (speech-based) and non-verbal(facial expressions, eye gaze, gesture, body pose, etc.) components.The verbal component contains semantic and affective information;accordingly, HRI work on the gesture component so far has focusedon rule-based (mapping words to gestures) and data-driven (deep-learning) approaches to generating speech-paired gestures basedon either semantics or the affective state. Consequently, most ges-ture systems are confined to producing either semantically-linkedor affect-based gesticures. This paper introduces an approach forenabling human-robot communication based on a theory-drivenapproach to generate speech-paired robot gestures using both se-mantic and affective information. Our model takes as input textand sentiment analysis, and generates robot gestures in terms oftheir shape, intensity, and speed.

  Access Paper or Ask Questions

Automatic Monitoring Social Dynamics During Big Incidences: A Case Study of COVID-19 in Bangladesh

Jan 31, 2021
Fahim Shahriar, Md Abul Bashar

Newspapers are trustworthy media where people get the most reliable and credible information compared with other sources. On the other hand, social media often spread rumors and misleading news to get more traffic and attention. Careful characterization, evaluation, and interpretation of newspaper data can provide insight into intrigue and passionate social issues to monitor any big social incidence. This study analyzed a large set of spatio-temporal Bangladeshi newspaper data related to the COVID-19 pandemic. The methodology included volume analysis, topic analysis, automated classification, and sentiment analysis of news articles to get insight into the COVID-19 pandemic in different sectors and regions in Bangladesh over a period of time. This analysis will help the government and other organizations to figure out the challenges that have arisen in society due to this pandemic, what steps should be taken immediately and in the post-pandemic period, how the government and its allies can come together to address the crisis in the future, keeping these problems in mind.

* Very minor change 

  Access Paper or Ask Questions

Bankruptcy prediction using disclosure text features

Jan 03, 2021
Sridhar Ravula

A public firm's bankruptcy prediction is an important financial research problem because of the security price downside risks. Traditional methods rely on accounting metrics that suffer from shortcomings like window dressing and retrospective focus. While disclosure text-based metrics overcome some of these issues, current methods excessively focus on disclosure tone and sentiment. There is a requirement to relate meaningful signals in the disclosure text to financial outcomes and quantify the disclosure text data. This work proposes a new distress dictionary based on the sentences used by managers in explaining financial status. It demonstrates the significant differences in linguistic features between bankrupt and non-bankrupt firms. Further, using a large sample of 500 bankrupt firms, it builds predictive models and compares the performance against two dictionaries used in financial text analysis. This research shows that the proposed stress dictionary captures unique information from disclosures and the predictive models based on its features have the highest accuracy.

* 36 pages, 14 figures 

  Access Paper or Ask Questions

Learning Neural Networks on SVD Boosted Latent Spaces for Semantic Classification

Jan 03, 2021
Sahil Sidheekh

The availability of large amounts of data and compelling computation power have made deep learning models much popular for text classification and sentiment analysis. Deep neural networks have achieved competitive performance on the above tasks when trained on naive text representations such as word count, term frequency, and binary matrix embeddings. However, many of the above representations result in the input space having a dimension of the order of the vocabulary size, which is enormous. This leads to a blow-up in the number of parameters to be learned, and the computational cost becomes infeasible when scaling to domains that require retaining a colossal vocabulary. This work proposes using singular value decomposition to transform the high dimensional input space to a lower-dimensional latent space. We show that neural networks trained on this lower-dimensional space are not only able to retain performance while savoring significant reduction in the computational complexity but, in many situations, also outperforms the classical neural networks trained on the native input space.

  Access Paper or Ask Questions

Causal Effects of Linguistic Properties

Oct 24, 2020
Reid Pryzant, Dallas Card, Dan Jurafsky, Victor Veitch, Dhanya Sridhar

We consider the problem of estimating the causal effects of linguistic properties on downstream outcomes. For example, does writing a complaint politely lead to a faster response time? How much will a positive product review increase sales? This paper focuses on two challenges related to the problem. First, we formalize the causal quantity of interest as the effect of a writer's intent, and establish the assumptions necessary to identify this from observational data. Second, in practice we only have access to noisy proxies for these linguistic properties---e.g., predictions from classifiers and lexicons. We propose an estimator for this setting and prove that its bias is bounded when we perform an adjustment for the text. The method leverages (1) a pre-trained language model (BERT) to adjust for the text, and (2) distant supervision to improve the quality of noisy proxies. We show that our algorithm produces better causal estimates than related methods on two datasets: predicting the effect of music review sentiment on sales, and complaint politeness on response time.

  Access Paper or Ask Questions

Cautious Monotonicity in Case-Based Reasoning with Abstract Argumentation

Jul 13, 2020
Guilherme Paulino-Passos, Francesca Toni

Recently, abstract argumentation-based models of case-based reasoning ($AA{\text -}CBR$ in short) have been proposed, originally inspired by the legal domain, but also applicable as classifiers in different scenarios, including image classification, sentiment analysis of text, and in predicting the passage of bills in the UK Parliament. However, the formal properties of $AA{\text -}CBR$ as a reasoning system remain largely unexplored. In this paper, we focus on analysing the non-monotonicity properties of a regular version of $AA{\text -}CBR$ (that we call $AA{\text -}CBR_{\succeq}$). Specifically, we prove that $AA{\text -}CBR_{\succeq}$ is not cautiously monotonic, a property frequently considered desirable in the literature of non-monotonic reasoning. We then define a variation of $AA{\text -}CBR_{\succeq}$ which is cautiously monotonic, and provide an algorithm for obtaining it. Further, we prove that such variation is equivalent to using $AA{\text -}CBR_{\succeq}$ with a restricted casebase consisting of all "surprising" cases in the original casebase.

  Access Paper or Ask Questions

Low Rank Fusion based Transformers for Multimodal Sequences

Jul 04, 2020
Saurav Sahay, Eda Okur, Shachi H Kumar, Lama Nachman

Our senses individually work in a coordinated fashion to express our emotional intentions. In this work, we experiment with modeling modality-specific sensory signals to attend to our latent multimodal emotional intentions and vice versa expressed via low-rank multimodal fusion and multimodal transformers. The low-rank factorization of multimodal fusion amongst the modalities helps represent approximate multiplicative latent signal interactions. Motivated by the work of~\cite{tsai2019MULT} and~\cite{Liu_2018}, we present our transformer-based cross-fusion architecture without any over-parameterization of the model. The low-rank fusion helps represent the latent signal interactions while the modality-specific attention helps focus on relevant parts of the signal. We present two methods for the Multimodal Sentiment and Emotion Recognition results on CMU-MOSEI, CMU-MOSI, and IEMOCAP datasets and show that our models have lesser parameters, train faster and perform comparably to many larger fusion-based architectures.

* ACL 2020 workshop on Second Grand Challenge and Workshop on Multimodal Language 

  Access Paper or Ask Questions

Learning What Makes a Difference from Counterfactual Examples and Gradient Supervision

Apr 20, 2020
Damien Teney, Ehsan Abbasnedjad, Anton van den Hengel

One of the primary challenges limiting the applicability of deep learning is its susceptibility to learning spurious correlations rather than the underlying mechanisms of the task of interest. The resulting failure to generalise cannot be addressed by simply using more data from the same distribution. We propose an auxiliary training objective that improves the generalization capabilities of neural networks by leveraging an overlooked supervisory signal found in existing datasets. We use pairs of minimally-different examples with different labels, a.k.a counterfactual or contrasting examples, which provide a signal indicative of the underlying causal structure of the task. We show that such pairs can be identified in a number of existing datasets in computer vision (visual question answering, multi-label image classification) and natural language processing (sentiment analysis, natural language inference). The new training objective orients the gradient of a model's decision function with pairs of counterfactual examples. Models trained with this technique demonstrate improved performance on out-of-distribution test sets.

  Access Paper or Ask Questions