On the Gap between Adoption and Understanding in NLP

Bianchi, Federico; Hovy, Dirk

doi:10.18653/v1/2021.findings-acl.340

Cited by 21 publications

(7 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They also seek to understand and explain how people behave, which means they also need to understand what drives the results of their statistical models. Interpretability allows researchers to scrutinize their models so that they might improve them and think about how well they might generalize to new contexts (Bianchi & Hovy, 2021). Improving interpretability can also improve fairness by allowing users (including regulatory bodies) to evaluate the model's strengths and failings in detail (Doshi-Velez et al, 2017;Rudin, 2019), and users generally trust models more when they understand them (Gilpin et al, 2018;Yeomans, Shah, et al, 2019).…”

Section: Feature-extraction Objectivesmentioning

confidence: 99%

A Practical Guide to Conversation Research: How to Study What People Say to Each Other

Yeomans,

Boland,

Collins

et al. 2023

Advances in Methods and Practices in Psychological Science

View full text Add to dashboard Cite

Conversation—a verbal interaction between two or more people—is a complex, pervasive, and consequential human behavior. Conversations have been studied across many academic disciplines. However, advances in recording and analysis techniques over the last decade have allowed researchers to more directly and precisely examine conversations in natural contexts and at a larger scale than ever before, and these advances open new paths to understand humanity and the social world. Existing reviews of text analysis and conversation research have focused on text generated by a single author (e.g., product reviews, news articles, and public speeches) and thus leave open questions about the unique challenges presented by interactive conversation data (i.e., dialogue). In this article, we suggest approaches to overcome common challenges in the workflow of conversation science, including recording and transcribing conversations, structuring data (to merge turn-level and speaker-level data sets), extracting and aggregating linguistic features, estimating effects, and sharing data. This practical guide is meant to shed light on current best practices and empower more researchers to study conversations more directly—to expand the community of conversation scholars and contribute to a greater cumulative scientific understanding of the social world.

show abstract

Section: Feature-extraction Objectivesmentioning

confidence: 99%

A Practical Guide to Conversation Research: How to Study What People Say to Each Other

Yeomans,

Boland,

Collins

et al. 2023

Advances in Methods and Practices in Psychological Science

View full text Add to dashboard Cite

show abstract

“…The Cost of the Hyperparameter Optimization. Although optimizing the hyperparameters of a topic model guarantees a fair comparison with other models, this approach is computationally expensive, possibly making it difficult to replicate the results (Bianchi and Hovy, 2021). In our work, we used BO because it is more efficient than other methods (Snoek et al, 2012;Bergstra and Bengio, 2012).…”

Section: Multi-objective Hyperparameter Optimization (Rs #1)mentioning

confidence: 99%

One Configuration to Rule Them All? Towards Hyperparameter Transfer in Topic Models using Multi-Objective Bayesian Optimization

Terragni¹,

Harrando²,

Lisena³

et al. 2022

Preprint

View full text Add to dashboard Cite

Topic models are statistical methods that extract underlying topics from document collections. When performing topic modeling, a user usually desires topics that are coherent, diverse between each other, and that constitute good document representations for downstream tasks (e.g. document classification). In this paper, we conduct a multi-objective hyperparameter optimization of three well-known topic models. The obtained results reveal the conflicting nature of different objectives and that the training corpus characteristics are crucial for the hyperparameter selection, suggesting that it is possible to transfer the optimal hyperparameter configurations between datasets.

show abstract

“…In keeping with the spirit of the first version of OCTIS, the framework extension is open-source and easily accessible, in order to guarantee researchers and practitioners a fairer, accessible and reproducible comparison between the models (Bianchi and Hovy, 2021). OCTIS 2.0 is available as extension of the original library, at the following link: https://github.com/mind-Lab/octis.…”

Section: Octis 20mentioning

confidence: 99%

Language Transfer for Identifying Diagnostic Paragraphs in Clinical Notes

Liello¹,

Uryupina²,

Moschitti³

2022

Proceedings of the Eighth Italian Conference on Computational Linguistics CliC-it 2021

View full text Add to dashboard Cite

The eighth edition of the Italian Conference on Computational Linguistics (CLiC-it 2021) was held at Università degli Studi di Milano-Bicocca from 26th to 28th January 2022.After the edition of 2020, which was held in fully virtual mode due to the health emergency related to Covid-19, CLiC-it 2021 represented the first moment for the Italian research community of Computational Linguistics to meet in person after more than one year of full/partial lockdown. Although the conference was held in dual mode, we strongly suggested the participants to attend it coming to Milan. Indeed, we received a strong feedback on this aspect from the community, which was eager to meet in person and enjoy both the scientific and social events together with the colleagues. In total, 99 participants registered to the conference benefiting from the early registration fee, 91 out of which expressed their intention to attend the event in person, which we consider as a very positive indication of enthusiasm from the community, given the uncertain situation due to the evolution of the pandemic in Italy.In total, we received 68 proposals, organized in the following specific tracks: Information Extraction,

show abstract

On the Gap between Adoption and Understanding in NLP

Cited by 21 publications

References 34 publications

A Practical Guide to Conversation Research: How to Study What People Say to Each Other

A Practical Guide to Conversation Research: How to Study What People Say to Each Other

One Configuration to Rule Them All? Towards Hyperparameter Transfer in Topic Models using Multi-Objective Bayesian Optimization

Language Transfer for Identifying Diagnostic Paragraphs in Clinical Notes

Contact Info

Product

Resources

About