Our team were excited to join this year's #Hacktoberfest with diverse #OpenSource contributions 🚀👏 From community project PRs and tackling data vis tool issues to making repos Hacktoberfest-eligible, the team even managed to plant multiple trees!💻🌲 hacktoberfest.com

Three more days left to apply for our Ph.D. position!

Time to get your applications in for the 3rd #DH #RSE Summer School, this year in 😍Edinburgh. We'll be there with a digital creativity themed day! Learn about software development for humanties research 1. Read the thingie (below) 2. Fill in the form 3. Press submit! It's easy!

Applications are now open for the #DigitalHumanities & #RSE Summer School 2024! 🌟 If you’re interested in becoming a Research Software Engineer in the humanities or integrating RSE practices into your research, apply now: cdcs.ed.ac.uk/DHRSE-Summer-S…

Big thanks to our team members federico nanni, Levan Bokeria, Dr Lydia France & @kasra_hosseini for helping to craft the course material for the Summer School in previous years! 🙌 You can check out the course details from last year here: github.com/CambridgeDH/DH…

Applications are now open for the Digital Humanities & Research Software Engineering Summer School 2024 (closing 17 May)! Find out more here: buff.ly/44fpG1w Edinburgh Centre for Data, Culture & Society The Alan Turing Institute Society of Research Software Engineering King's Digital Lab (@kingsdigitallab.bsky.social)

Now it's website-official! Announcing our new AHRC-funded Data/Culture project at The Alan Turing Institute where I get to extend the mission of Living with Machines with amazing colleagues Pieter Francois Katie McDonough Daniel 🤖💀 Wilson @dan9000.bluesky.social David Beavan federico nanni Nilo Pedrazzini turing.ac.uk/research/resea…

Applications close on Friday! Early career digital humanists and adjacent colleagues - don't miss out on this brilliant opportunity. Learn more: cdcs.ed.ac.uk/DHRSE-Summer-S…

Can Vision Language Models aid in the detection and description of phenotypes? With federico nanni, we sought to test this for Drosophila wing images. Preprint here: biorxiv.org/content/10.110…

The Turing Way by ⁦The Alan Turing Institute⁩ mentioned by a presentation at #aiucd2024 Great papers by researchers ⁦Università di Bologna⁩ https//centri.unibo.it/dharc/en

We have an opening for a PhD on "Large Language Models for a personalised access to large cultural data" See MSCA COFUND - Futuredata4EU Call, page 69: site.unibo.it/futuredata4eu/… Competitive EU salary for 3y. Deadline June 28th. Francesca Tomasi Silvio Peroni /DH.arc #GLAM #LLMs #DH

⚽️ We're excited to share the AIrsenal predictions for the #Euros2024! This side project by our team uses #MachineLearning and #AI to predict #football outcomes Currently, France 🇫🇷 are favorites, with England 🏴󠁧󠁢󠁥󠁮󠁧󠁿 in 2nd spot and Scotland 🏴󠁧󠁢󠁳󠁣󠁴󠁿 with a slim chance down in 20th.

New paper out! Very excited that we’re able to share STAR: SocioTechnical Approach to Red Teaming Language Models. We've made some methodological advancements focusing on human red teaming for ethical and social harms. 🧵Check out arxiv.org/abs/2406.11757

Excited to be sharing the results of some of our recent work at The Alan Turing Institute @turingpubpol! ✨ We explore how effective LLMs are for disinformation operations, presenting a new eval and experiments around election disinformation. arxiv.org/abs/2408.06731

🚀 Want to query LLMs at scale? 🤔 Wish there was a package that could optimise rate/token limits for popular API providers? And could even hook into self-hosted open source models? We’re releasing prompto: a library designed to support reproducible LLM eval experiments.🧪

💻 Frontier performance on local compute Our Project t0 team has shown that small, open-weight language models can deliver near-frontier reasoning performance when applied to focused, domain-specific tasks.

Retrieval-Augmented Reasoning with Lean Language Models Great paper showing how to fuse RAG and reasoning into a single small-footprint language model. Distillation works if done correctly. Very exciting results! Here are my notes: