Interesting findings--e.g. 'surfacing disagreement through rater diversity'... #ethics #AI #LLMs #data #tech #research

This talk was an absolute highlight of my NeurIPS Conference experience! Really thought provoking points were made about how we go about our data collection and analysis (with a strong emphasis on fairness and responsibility)

この中でおすすめしている招待講演のスライドが今日公開されたようです（すごくタイムリー）

Turns out that diversity among the raters annotating #data matters a lot. #ethics #AI #tech #research #business

A big bunch of nails in the coffin of the idea that open source AI models are more dangerous than closed ones.
And a huge warning to regulators who are on the way to throw the AI baby with the regulatory bath water.

What has been happening since the release of Llama-2 is an

Blue note CD no. 2 TECHNICALLY ACCEPTABLE is now on all platforms and available from all outlets! I'll be talking about this music in my feeds for the next little bit. Thanks for listening.

Kinda amazing to see a historical overview of myself like this. Thank you burning ambulance dot substack dot com

Here are the slides of my NeurIPS Conference 2023 #Keynote : The Many Faces of Responsible AI - discussing topics of #dataquality #dataexcellence #dataforai slideshare.net/laroyo/neurips…

If you’re working on LLM safety, check out SafetyPrompts.com!

SafetyPrompts.com is a catalogue of open datasets for evaluating and improving LLM safety. I started building this over the holidays, and I know there are still datasets missing, so I need your help 🧵

15,000 people in one room at #NeurIPS 2023 for a keynote is remarkable. Of course it’s on the topic of LLMs (Large Language Models). Dr. Lora Aroyo talks about the many faces of responsible AI. See her three conclusion slides in the pictures below. #AI #ArtificialIntelligence

I took notes for Lora Aroyo @neurips2023 's invited talk today at NeurIPS Conference: 'The Many Faces of Responsible AI'
theyre pretty detailed l o l 😅 #NeurIPS2023
diamond-tilapia-430.notion.site/Lora-Aroyo-The…

Yes.

Many others have said it before me and likely better than me, through their research and organizing, such as Diyi Yang Dirk Hovy Daniel Hershcovich Sunipa Dev Rida Qadri Luciana Benotti Ben Hutchinson Fernando Diaz Ife Adebara, and many others I'm missing here...

Congratulations to Jeff Dean, Greg Corrado, & co-authors of the paper “Distributed Representations of Words and Phrases and their Compositionality”, for winning the #NeurIPS2023 Test of Time Award! This prize recognizes a highly impactful paper published at NeurIPS 10 years ago.

Thanks for taking these notes!! Really nice capture

Lia Coleman (they/them) Lora Aroyo @neurips2023 NeurIPS Conference these notes are great! thanks for getting them down, and thanks Lora Aroyo @neurips2023 for such a inspiring talk

Stop by the #NeurIPS2023 Google booth today at 9:15am where Lora Aroyo @neurips2023, Vinodkumar Prabhakaran, & Alicia Parrish will introduce the DICES dataset, a shared resource and benchmark that respects diverse perspectives during safety evaluation of conversational AI systems. github.com/google-researc…

Don't miss out on the chance to visit the #NeurIPS2023 Google booth today at 9:15am! Join Lora Aroyo @neurips2023, Vinodkumar Prabhakaran, & Alicia Parrish as they unveil the groundbreaking DICES dataset. This shared resource and benchmark ensures the inclusion of diverse perspectives in the safety

The Adversarial Nibbler Challenge is a prompt hacking competition for implicit adversarial prompts with the goal of improving generative text-to-image model safety one bite at a time. At 12:45pm today, visit the #NeurIPS2023 Google booth where Lora Aroyo @neurips2023 will talk all about it!

Don't miss Lora Aroyo @neurips2023 give the #NeurIPS2023 opening Keynote talk, “The Many Faces of Responsible AI”, in which she'll discuss the need for culturally-aware, society-centered research on the impact of data quality and diversity for training and evaluation of ML models. Happening now!

Drop by the Google booth at 9:15am today where Lora Aroyo @neurips2023, Meg Risdal & Omar Benjelloun will introduce Croissant 🥐—in collaboration with MLCommons & others—a format for ML datasets with the goal of making datasets interoperable across repos & platforms. github.com/mlcommons/croi…