Data-centric NLP in the era of LLMs

https://is1-ssl.mzstatic.com/image/thumb/Podcasts116/v4/ad/c3/3e/adc33eac-7ce9-561d-7041-b2362e3709cc/mza_16967565627371900312.jpg/600x600bb.jpg

D4 Data Podcast

Deepak John Reji

56 episodes

2 weeks ago

Join us on the D4 Data Channel podcast as we delve into the ethical and societal implications of AI and other digital technologies and their role in promoting Sustainability. In each episode, we invite professionals to share their stories and expertise in relevant fields. In addition to technical topics, we also cover personal development, and how individuals can leverage technology to advance their careers and build their personal brands.

Technology

RSS

All content for D4 Data Podcast is the property of Deepak John Reji and is served directly from their servers with no modification, redirects, or rehosting. The podcast is not affiliated with or endorsed by Podjoint in any way.

Technology

https://d3t3ozftmdmh3i.cloudfront.net/production/podcast_uploaded_episode/31417200/31417200-1674210176794-19a408ce0c957.jpg

Data-centric NLP in the era of LLMs

D4 Data Podcast

32 minutes 4 seconds

2 years ago

Data-centric NLP in the era of LLMs

In this episode, we dive into the world of data-centric AI and NLP, exploring the ethical considerations, limitations, and trade-offs that come with this approach. We start off discussing data-centric AI, and the ethical considerations that come with it in the context of NLP. We also delve into the limitations of data-centric approaches in NLP, and the trade-offs between using a larger dataset versus a curated, smaller dataset. We also touch on the problem of concept drift and the potential for generating biased or misleading data while augmenting data. We also discuss the importance of ensuring the quality of labels obtained through crowdsourcing and how to achieve that.

Daniel Vila Suero shared with us why Argilla is open source and how Argilla is adapting to the rapidly changing landscape in the AI industry. Join us as we explore the challenges and opportunities in data-centric NLP and discover how Argilla is pushing the boundaries of what's possible with these models.

Daniel Vila Suero is the CEO and co-founder of Argilla. During his PhD thesis on Artificial Intelligence (UPM, 2016), Daniel designed, developed and deployed the end-to-end framework powering the datos.bne.es service from the National Library of Spain. Daniel left academia in 2017 to found Argilla.

https://www.argilla.io/blog/recognai-...

Argilla: https://www.argilla.io/