PAVIS and former members celebrating the coming holidays and a year full of successes! Well done people, you are an amazing team! IIT

Attention has been the key component for most advances in LLMs, but it can’t scale to long context. Does this mean we need to find an alternative? Presenting Titans: a new architecture with attention and a meta in-context memory that learns how to memorize at test time. Titans

🚀 Announcing ASAP: asap-seminar.github.io! A fully virtual seminar bridging theory, algorithms, and systems to tackle fundamental challenges in Transformers. Co-organized by Simran Arora Xinyu Yang Han Guo Our first speaker: Alex Wang on Test-time Regression

If you want to achieve AGI, you need to solve continual learning! CoLLAs 2025 is the premier venue for continual learning research! Check this blog post to read more about CoLLAs 2025!

Working in #Multimedia, #ComputerVision, #AI? The Deadline for #ACMMM2025 is approaching fast! Don't forget to follow the new official profile for updates! x.com/acmmm_social

The first edition of Learning Over Time (LOT) Spring School is starting! Stefano Melacci kicking off the school at the wonderful Certosa di Pontignano, Italy Organized by Stefano Melacci, Vincenzo Lomonaco , Andrea Cossu , Alessandro Betti , and myself cai.diism.unisi.it/school/

Art, nature and Artificial Intelligence at LoT Spring school which just started. With 50+ students, we are fully booked and eager to learn (over time). Stefano Melacci Vincenzo Lomonaco Matteo Tiezzi Alessandro Betti, Marco Gori

#LOT spring school we continue our incredible lineup of speakers with Alexei Efros and the talk titled "Test-Time Training: Toward a Post-Dataset World", Stefano Melacci Andrea Cossu Vincenzo Lomonaco Alessandro Betti cai.diism.unisi.it/school/

#LOT Spring school Extremely honoured to introduce the second talk of the day, by the great Antonio Orvieto, on "New Avenues in Long-Sequence Processing without Attention" cai.diism.unisi.it/school/ Stefano Melacci Andrea Cossu Vincenzo Lomonaco Alessandro Betti

#LOT Spring school Battista Biggio presenting the lecture "Machine Learning Security: Lessons Learned and Future Challenges" Andrea Cossu Vincenzo Lomonaco Stefano Melacci Alessandro Betti

#LOT Spring School Elisa Ricci on "Beyond Zero-Shot Generalization: Strengthening Vision-Language Models with Adaptation & Personalization" cai.diism.unisi.it/school/ Stefano Melacci Vincenzo Lomonaco Andrea Cossu

#LOT very interesting interactive lecture by Claudio Gallicchio "Reservoir Computing Methods for Time Series Classification" Andrea Cossu Vincenzo Lomonaco Stefano Melacci

#LOT startup day with ContinualIST (Vincenzo Lomonaco), Motus ML (Giacomo Ziffer ) and FocoosAI (Fabio Cermelli) Andrea Cossu Stefano Melacci Alessandro Betti

1/9 There is a fundamental tradeoff between parallelizability and expressivity of Large Language Models. We propose a new linear RNN architecture, DeltaProduct, that can effectively navigate this tradeoff. Here's how!

Julien Siems leloy! Jyo Pari In our DeltaProduct work we also add a bit of theory to DeltaNet, showing that it can solve Dihedral groups, which are the groups of symmetries of regular polygons, with only two layers. This includes S3 (symmetries of the equilateral triangle).

🚨 Our #survey "Back to recurrent processing at the crossroad of #transformers and state-space models" is out in Nature Machine Intelligence! 🔗 nature.com/articles/s4225… Read it here: rdcu.be/el85K Stefano Melacci Marco Gori (M. Gori)

A great thread that showcases the main intuitions behind our #FovEx paper Paper: 🔗 link.springer.com/article/10.100… Technical Report: 🔗arxiv.org/abs/2408.02123 Github: 📷 github.com/mahadev1995/Fo…