Dirk Hovy's guest appearances

Dirk Hovy

NLP Highlights

Let's Chat Ethics

The Consumer Researcher

Sebastian Ruder

Ines Montani

Noah Smith

Michael Collins

Brendan O'Connor

Victor Sanh

Sam Bowman

Nathan Schneider

Matt Gardner

Chris Dyer

John Hewitt

Dan Roth

Leon Derczynski

Zhou Yu

Jon Clark

OwlTail

Older and wiser, Oriana and Amanda are back to chatting ethics after a hiatus! In this episode, we dive into the world of AI and the technology behind it. Meet ChatGPT, a large language model developed by OpenAI, and learn about its capabilities, limitations, and potential impact on our daily lives. From language generation to answering complex questions, we'll discover how ChatGPT works and how it's being used to enhance human capabilities. Join us as we engage in a conversation with Prof. Dirk Hovy to understand the ethical implications and the future of this rapidly advancing technology. Get ready to be amazed and informed as we explore the fascinating world of AI.  (*)This description was written using chatGPT and some human editing skills. 

Older and wiser, Oriana and Amanda are back to chatting ethics after a hiatus! In this episode, we dive into the world of AI and the technology behind it. Meet ChatGPT, a large language model developed by OpenAI, and learn about its capabilities, limitations, and potential impact on our daily lives. From language generation to answering complex questions, we'll discover how ChatGPT works and how it's being used to enhance human capabilities. Join us as we engage in a conversation with <a href="https://faculty.unibocconi.eu/dirkhovy/" rel="nofollow" target="_blank">Prof. Dirk Hovy</a> to understand the ethical implications and the future of this rapidly advancing technology. Get ready to be amazed and informed as we explore the fascinating world of AI.  (*)This description was written using chatGPT and some human editing skills.

We are back! Let's Chat chatGPT with Dirk Hovy

Shiri Melumad

Jeff Inman

Wordify is a new methodological tool for discovering and differentiating consumer vocabularies. Listen to editors Markus Giesler and Bernd Schmitt discuss Wordify with the method's creators and authors of the associated JCR article, Dirk Hovy, Shiri Melumad, and Jeff Inman.

Wordify is a new methodological tool for discovering and differentiating consumer vocabularies. Listen to editors Markus Giesler and Bernd Schmitt discuss Wordify with the method's creators and authors of the associated JCR article, Dirk Hovy, Shiri Melumad, and Jeff Inman.

Author Interview: Dirk Hovy, Shiri Melumad, and Jeff Inman, authors of Wordify

EMNLP 2018 paper by Dirk Hovy and Tommaso Fornaciari. https://www.semanticscholar.org/paper/Improving-Author-Attribute-Prediction-by-Linguistic-Hovy-Fornaciari/71aad8919c864f73108aafd8e926d44e9df51615In this episode, Dirk Hovy talks about natural language as social phenomenon which can provide insights about those who generate it. For example, this paper uses retrofitted embeddings to improve on two tasks: predicting the gender and age group of a person based on their online reviews. In this approach, authors embeddings are first generated using Doc2Vec, then retrofitted such that authors with similar attributes are closer in the vector space. In order to estimate the retrofitted vectors for authors with unknown attributes, a linear transformation is learned which maps Doc2Vec vectors to the retrofitted vectors. Dirk also used a similar approach to encode geographic information to model regional linguistic variations, in another EMNLP 2018 paper with Christoph Purschke titled “Capturing Regional Variation with Distributed Place Representations and Geographic Retrofitting” [link: https://www.semanticscholar.org/paper/Capturing-Regional-Variation-with-Distributed-Place-Hovy-Purschke/6d9babd835d0cdaaf175f098bb4fd61fd75b1be0].

EMNLP 2018 paper by Dirk Hovy and Tommaso Fornaciari. <a href="https://www.semanticscholar.org/paper/Improving-Author-Attribute-Prediction-by-Linguistic-Hovy-Fornaciari/71aad8919c864f73108aafd8e926d44e9df51615" rel="nofollow" target="_blank">https://www.semanticscholar.org/paper/Improving-Author-Attribute-Prediction-by-Linguistic-Hovy-Fornaciari/71aad8919c864f73108aafd8e926d44e9df51615</a> In this episode, Dirk Hovy talks about natural language as social phenomenon which can provide insights about those who generate it. For example, this paper uses retrofitted embeddings to improve on two tasks: predicting the gender and age group of a person based on their online reviews. In this approach, authors embeddings are first generated using Doc2Vec, then retrofitted such that authors with similar attributes are closer in the vector space. In order to estimate the retrofitted vectors for authors with unknown attributes, a linear transformation is learned which maps Doc2Vec vectors to the retrofitted vectors. Dirk also used a similar approach to encode geographic information to model regional linguistic variations, in another EMNLP 2018 paper with Christoph Purschke titled “Capturing Regional Variation with Distributed Place Representations and Geographic Retrofitting” [link: <a href="https://www.semanticscholar.org/paper/Capturing-Regional-Variation-with-Distributed-Place-Hovy-Purschke/6d9babd835d0cdaaf175f098bb4fd61fd75b1be0" rel="nofollow" target="_blank">https://www.semanticscholar.org/paper/Capturing-Regional-Variation-with-Distributed-Place-Hovy-Purschke/6d9babd835d0cdaaf175f098bb4fd61fd75b1be0</a>].

76 - Increasing In-Class Similarity by Retrofitting Embeddings with Demographics, with Dirk Hovy