🦕The 19th conference on Neurosymbolic AI will be in beautiful Santa Cruz (CA, USA), September 8-10, 2025! CFP is now out (See thread) 🚨 Paper deadline: Feb 28 (abstract), March 7 (full) #neurosymbolic #NeSy2025

Good morning from day 2 at NeSy in Santa Cruz! Starting with a nice oral session, such as studying how well neural networks learn boolean functions (with some surprising results!) @luislamb and verifying NeSy models for robustness

The (iconic) school bus to NeSy https://t.co/laA175VQfa

X loops back into the real world... via my colleague at NeSy!

Do world models need objects? @tkipf made a strong case today at @nesyconf from objects as the symbols of the physical world to SlotFormer, enabling compositional video dynamics prediction from just a single frame. Exciting directions for object-centric generative models.

@e_giunchiglia @tkipf Object representations can be used in different settings: - Videos - 3D modelling - Robotic representations This especially helps in the low-data regime!

@e_giunchiglia It's time for @tkipf 's keynote! Does object-centric representations still matter in the age of Generative AI where Transformers are key? Yes - But it's not part of the architecture:

@e_giunchiglia @guyvdb Behind all of these very nice methods are one central trick... Circuits! ➕✖️ These are tractable generative neural networks 😍

Just by adding the DFA that constraints the reasoning to include the number at least once, it improved the performance. This is a good illustration of how simple rules can significantly improve the performance of LLMs. #Neurosymbolic #Nesy

@e_giunchiglia @guyvdb Did you know Neurosymbolic methods can beat 'the other's turf'? Just use a 100x smaller LM with symbolic constraints integrated via Ctrl-G to wipe the floor with GPT!🧹 https://t.co/1cZJgSTc92 @guyvdb

@e_giunchiglia @guyvdb How can reverend Bayes help us to incorporate constraints? With NeSy of course 👀 With applications in non-toxic LLM generation and safe AI driving! @guyvdb

@e_giunchiglia Now, @guyvdb is giving the opening keynote arguing why symbolic AI is still relevant in the age of LLMs... With the help of Shrek!

Good morning from NeSy! We are very happy to welcome you here in Santa Cruz, and are glad to announce the conference is growing 📈 Check out this thread below for live updates from the conference 🦕

Our second keynote speaker is @tkipf, who will discuss object-centric representation learning! Do objects need a special treatment for generative AI and world models? 🤔 We will hear on Monday!

It is almost time to welcome you all in Santa Cruz! 🦕 We will start with an exciting and timely keynote by @guyvdb on "Symbolic Reasoning in the Age of Large Language Models" 👀

🚀 Launching at #NeSy2025: the OPTIMAS × NeSy Hackathon 3 Challenges • 2 Phases • $45K in Prizes + Job Opportunity Build neurosymbolic reasoning systems (NL→Ontology, Causal Impact, Ontology Evolution). Teams 1–5. Sept 9 (3–5 PM). 🔗 https://t.co/mUsiyPiV36 #NeuroSymbolic #AI

Neurosymbolic AI is very much concerned with how to do this wiring: adding prior knowledge / constraints to the network architecture or loss function (instead of chain of thought):

#Nesy2025 schedule is out! Keynote speakers include @guyvdb @tkipf @dlmcguinness and @GaryMarcus. I'm looking forward to seeing what everyone's cooking and meeting everyone! @nesyconf https://t.co/5AS6shaNoH

🎓 Travel grants for US PhD students & ECRs — Apply by July 20 🔗 https://t.co/P3SGhZvkN2 💡 Early registration ends Aug 7 🔗 https://t.co/QFJr1CNByU #NeSy2025 #NeuroSymbolicAI

🚨 First Call for Participation – NeSy 2025 📍 Sept 8–10 | Santa Cruz, CA Join the longest-running conference on neurosymbolic AI! Our keynote speakers: @guyvdb , @tkipf , @dlmcguinness , @GaryMarcus More info 👇

Openreview: https://t.co/fiUshOvVhI CFP: