Jeff Dalton

I’ve spent my career studying the science of search. The new generation of LLM models asks us to fundamentally reimagine everything. Here's what it mean's for AI Coaching at Valence: valence.co/blog/breaking-…

The next post in my #SWRL2025 series is up on Linkedin - linkedin.com/posts/jeffjdal… It covers the arrival to Melbourne and the opening gathering setting the stage for the week. Coffee pictures with @lauradietz99.

This week I’m in Australia for the 4th Strategic Workshop on Information Retrieval (sites.google.com/view/swirl2025…). I’ll be posting with some of the highlights, big questions, new breakthroughs, and more. See post details on Linkedin - linkedin.com/posts/jeffjdal…

The future belongs to those who do evals

Come and chat about our #KDD2024 poster (#99) about lessons learned during the development of GRILLBot, our #AlexaPrize conversational task assistant ☀️ w/ @JeffD @carlos_gemmell @iain_with_2is @FedingoRoss91 @grill_lab_ai @EdinburghNLP Preprint: arxiv.org/pdf/2402.07647

Excited that our paper "GRILLBot In Practice: Lessons and Tradeoffs Deploying Large Language Models for Adaptable Conversational Task Assistants" will appear at #KDD2024 - including many @AmazonScience #AlexaPrize insights from @grill_lab_ai! Preprint: arxiv.org/pdf/2402.07647

We are thrilled to announce we have closed a £6m seed round led by @HoxtonVentures . To shrink AI models to solve the world’s hardest enterprise problems. This follows our £1m pre-seed in August led by @CreatorFund sifted.eu/articles/malte…

@SnowflakeDB is open sourcing the best embedding models in the world! 🚀🚀 They are now available open source in @huggingface We are releasing it under the Apache 2 license so that it is easy for the OSS community to experiment with them freely.🎁🎁 These impressive models are the result of @Neeva's expertise in search, state-of-the-art research, and @SnowflakeDB ’s steadfast commitment to excellence in AI. It’s incredible to see what the team has built, and even better to delight our customers day after day with our latest innovations.

Research rethinking a search pipeline leveraging passage-level feedback and entity representations. @grill_lab_ai. Also available on arxiv: arxiv.org/abs/2306.17082

Congratulations 🎊🎉 to all members! More exciting work on LLMs and conversational search coming up. Finally, we are hiring PhD students and postdocs! Get in touch with PI Jeff Dalton (jeff.dalton@ed.ac.uk) to discuss the positions. (6/6)

Visiting Professor @JeffD of @InfAtEd & our #AI Group will discuss real-world challenges deploying generative LLMs in his #ECIR2024 Industry Day keynote (11 AM GMT): "Generative AI in Finance: Automatic, Topic-based Summaries for Earnings Call Transcripts" bloom.bg/3ITuiQP

And for a bit of throwback, my first talk at a research conference was at the first ECIR Industry Day in 2008. From my days working on search quality @GlobalSpecLLC before starting a PhD @manningcics later in the Fall. From the Wayback Machine: web.archive.org/web/2009021402…

Today @ 11, I'm excited to speak at the #ecir2024 Industry Day giving a keynote on "Generative AI in Finance: Automatic, Topic-based Summaries for Earnings Call Transcripts" with @TechAtBloomberg.

My group, @grill_lab_ai at @InfAtEd has openings for multiple postdocs and PhD students to work on the future of interactive assistants, generative retrieval, and Large Language Models (LLMs). We're at #ecir2024, so please come chat.

Exciting new work on scaling creation of personalized search sessions with LLMs. Congrats team! @grill_lab_ai.

@julia_kiseleva It was very tempting! Logistics didn't quite work this year. Sorry to miss it!

@JeffD I am missing so much your food map :-) why haven’t you came to Malta? 😀

I look forward to seeing you at #ecir2024 in Glasgow soon! As my current base, I know it well. I'm happy to share two food maps: - A top 50 Food & Drink list just for ECIR maps.app.goo.gl/Ku8gGABAZfwVAL… - My full list maps.app.goo.gl/hTZRzVaHWam3CQ… Research welcome!

A lot of fun being @cikm2023