Identifying Exaggerated Language

Kong, Lingqi; Li, Chuanyi; Ge, Jidong; Luo, Bin; Ng, Vincent

doi:10.18653/v1/2020.emnlp-main.571

Cited by 7 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…with rhythm and repetition. Studies include the detection of repetitive figures (Dubremetz and Nivre, 2018), exaggeration (Troiano et al, 2018;Kong et al, 2020) and of syntax figures (Al Khatib et al, 2020).…”

Section: Mapping Persuasion Modellingmentioning

confidence: 99%

Modelling Persuasion through Misuse of Rhetorical Appeals

Паули¹,

Derczynski²,

Assent³

2022

Proceedings of the Second Workshop on NLP for Positive Impact (NLP4PI)

View full text Add to dashboard Cite

It is important to understand how people use words to persuade each other. This helps understand debate, and detect persuasive narratives in regard to e.g. misinformation. While computational modelling of some aspects of persuasion has received some attention, a way to unify and describe the overall phenomenon of when persuasion becomes undesired and problematic, is missing. In this paper, we attempt to address this by proposing a taxonomy of computational persuasion. Drawing upon existing research and resources, this paper shows how to re-frame and re-organise current work into a coherent framework targeting the misuse of rhetorical appeals. As a study to validate these re-framings, we then train and evaluate models of persuasion adapted to our taxonomy. Our results show an application of our taxonomy, and we are able to detecting misuse of rhetorical appeals, finding that these are more often used in misinformative contexts than in true ones. Fallacy of PathosAppeal to Fear: "'Without this additional insurance, you could find yourself broke and homeless" Appeal to Pity: "I know I missed assignments, but if you fail me, I will lose my financial aid and have to drop out." Appeal to Popularity: "Nine out of ten shoppers have switched to Blindingly-Bright-Smile Toothpaste." Fallacy of EthosFalse authority:"Dr. X is an engineer, and he doesn't believe in global warming." Ad Hominem:"Why should we think a candidate who recently divorced will keep her campaign promises?" Name-calling:"These rabble-rousers are nothing but feminazis." Fallacy of LogosFalse dilemma:"Either we pass this ordinance or there will be rioting in the streets" Circular argument:"This legislation is sinful because it is the wrong thing to do." Red Herring or Smoke Screen:"My opponent says I am weak on crime, but I have been one of the most reliable participants in city council meetings.

show abstract

Section: Mapping Persuasion Modellingmentioning

confidence: 99%

Modelling Persuasion through Misuse of Rhetorical Appeals

Паули¹,

Derczynski²,

Assent³

2022

Proceedings of the Second Workshop on NLP for Positive Impact (NLP4PI)

View full text Add to dashboard Cite

show abstract

“…Hyperbole Corpus Troiano et al (2018) built the HYPO dataset consisting of 709 hyperbolic sentences with human-written paraphrases and lexically overlapping nonhyperbolic counterparts. Kong et al (2020) also built a Chinese hyperbole dataset with 2680 hyperboles. Our HYPO-L and HYPO-XL are substantially larger than HYPO and we hope they can facilitate computational research on hyperbole detection and generation.…”

Section: Related Workmentioning

confidence: 99%

“…Hyperbole detection is a supervised binary classification problem where we predict whether a sentence is hyperbolic or not (Kong et al 2020). We fine-tune BERT (Devlin et al 2019) on the hyperbole detection dataset HYPO (Troiano et al 2018).…”

Section: Appendix a Hyperbole Corpus Collectionmentioning

confidence: 99%

“…The second one (from the Minimal Unit Corpus) is a sentence from Wikipedia that includes the minimal hyperbolic span, but with a literal connotation. Kong et al (2020) have demonstrated empirically on a Chinese hyperbole detection dataset HYPO-cn that deep learners can offer better performance than traditional algorithms with hand-crafted features. As a result, we fine-tune BERT-base model on HYPO.…”

Section: Appendix a Hyperbole Corpus Collectionmentioning

confidence: 99%

“…Although hyperbole is considered as the second most frequent rhetoric device (Kreuz and Roberts 1993), it has received less empirical attention in the NLP community. While hyperbole detection has already been investigated by Troiano et al (2018) and Kong et al (2020), there has been no study on automatic hyperbole generation and we aim to fill this research gap. From a syntactic point of view, Claridge (2010) classifies hyperbole into word-level, phrase-level and clause-level types, and conclude that the former two types are more common in English.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MOVER: Mask, Over-generate and Rank for Hyperbole Generation

Zhang¹,

Wan²

2021

Preprint

View full text Add to dashboard Cite

Despite being a common figure of speech, hyperbole is underresearched with only a few studies addressing its identification task. In this paper, we introduce a new task of hyperbole generation to transfer a literal sentence into its hyperbolic paraphrase. To tackle the lack of available hyperbolic sentences, we construct HYPO-XL, the first large-scale hyperbole corpus containing 17,862 hyperbolic sentences in a non-trivial way. Based on our corpus, we propose an unsupervised method for hyperbole generation with no need for parallel literal-hyperbole pairs. During training, we fine-tune BART to infill masked hyperbolic spans of sentences from HYPO-XL. During inference, we mask part of an input literal sentence and over-generate multiple possible hyperbolic versions. Then a BERT-based ranker selects the best candidate by hyperbolicity and paraphrase quality. Human evaluation results show that our model is capable of generating hyperbolic paraphrase sentences and outperforms several baseline systems. 1

show abstract

A Survey on Automatic Generation of Figurative Language: From Rule-based Systems to Large Language Models

Lai,

Nissim

2024

ACM Comput. Surv.

View full text Add to dashboard Cite

Figurative language generation (FLG) is the task of reformulating a given text to include a desired figure of speech, such as a hyperbole, a simile, and several others, while still being faithful to the original context. This is a fundamental, yet challenging task in Natural Language Processing (NLP), which has recently received increased attention due to the promising performance brought by pre-trained language models. Our survey provides a systematic overview of the development of FLG, mostly in English, starting with the description of some common figures of speech, their corresponding generation tasks and datasets. We then focus on various modelling approaches and assessment strategies, leading us to discussing some challenges in this field, and suggesting some potential directions for future research. To the best of our knowledge, this is the first survey that summarizes the progress of FLG including the most recent development in NLP. We also organize corresponding resources, e.g., paper lists and datasets, and make them accessible in an open repository. We hope this survey can help researchers in NLP and related fields to easily track the academic frontier, providing them with a landscape and a roadmap of this area.

show abstract

Identifying Exaggerated Language

Cited by 7 publications

References 13 publications

Modelling Persuasion through Misuse of Rhetorical Appeals

Modelling Persuasion through Misuse of Rhetorical Appeals

MOVER: Mask, Over-generate and Rank for Hyperbole Generation

A Survey on Automatic Generation of Figurative Language: From Rule-based Systems to Large Language Models

Contact Info

Product

Resources

About