RSTLess group

🎉 Two papers at #SIGIR2026! 🐱 CATS: Clustering + Thompson Sampling for smarter negative mining in RecSys. +20% NDCG@10. ✂️ Load-sensitive Selective Pruning: dynamic embedding dimensionality for dense retrieval under load. +400% queries served. See you in Melbourne! #IR #RecSys

Giulia Iadisernia is a first-year PhD student in Data Science at Sapienza, supervised by Fabrizio Silvestri a. Her research focuses on the #trustworthiness and #explainability of LLMs. She also collaborates with the Applied Research Team at the Bank of Italy.

Alessia Borghini is a first-year PhD student in the National Doctorate in AI supervised by Fabrizio Silvestri. Her research focuses on Mechanistic Interpretability for #AISafety, aiming to understand LLMs and improve alignment and reliability. Fell free to reach out!!

New paper from @wanifarooq848 on Vision Language Models! Read it on arxiv! #RSTLess #Sapienza

Alessio Borgi is a PhD student in Graph Neural Networks & Generative AI—bridging geometry, topology, and diffusion models for robotics, vision, and biomedical AI. Feel free to contact him for collaborations. #AI #GNN #GenerativeAI #Robotics #RSTLess #Sapienza

Gavriel Di Nepi is a first-year PhD student in Data Science at Sapienza under the supervision of prof. @fabreetseo . His research focuses on human-inspired memory for #AI to boost performance and reasoning, in collaboration with @bakerhughesco. #RSTLess #Sapienza

Meet Maria Vittoria Vestini, 1st-year PhD student in Data Science at Sapienza University of Rome, supervised by @fabreetseo. Her research explores interactions and learning dynamics between large and small language models. Learn more: rstless.it #RSTLess #Sapienza

By combining the expressive power of sheaves with a principled treatment of directionality, our method achieves consistent relative accuracy improvements over prior methods.

Directional Sheaf Hypergraph Networks: Unifying Learning on Directed and Undirected Hypergraphs accepted at #ICLR2026!! Work done by Emanuele Mule, Stefano Fiorini, @AntonioPuri00, @_FedeSiciliano_ , Stefano Coniglio and @fabreetseo. Paper: arxiv.org/pdf/2510.04727

- We introduce the notion of Directed Cellular Sheaves over directed hypergraphs - From this construction, we derive the Directed Sheaf Hypergraph Laplacian, a complex-valued Hermitian operator that unifies and generalizes existing graph and hypergraph convolutional operators.

It presents theoretical and empirical results answering to the question: "𝐶𝑎𝑛 𝑀𝑎𝑗𝑜𝑟𝑖𝑡𝑦 𝑉𝑜𝑡𝑖𝑛𝑔 𝑎𝑐ℎ𝑖𝑒𝑣𝑒 𝑡ℎ𝑒 𝑡ℎ𝑒𝑜𝑟𝑒𝑡𝑖𝑐𝑎𝑙𝑙𝑦 𝑜𝑝𝑡𝑖𝑚𝑎𝑙 𝑙𝑎𝑏𝑒𝑙 𝑒𝑠𝑡𝑖𝑚𝑎𝑡𝑖𝑜𝑛 𝑒𝑟𝑟𝑜𝑟 𝑎𝑛𝑑 𝑢𝑛𝑑𝑒𝑟 𝑤ℎ𝑎𝑡 𝑐𝑜𝑛𝑑𝑖𝑡𝑖𝑜𝑛𝑠?"

𝐓𝐡𝐞 𝐌𝐚𝐣𝐨𝐫𝐢𝐭𝐲 𝐕𝐨𝐭𝐢𝐧𝐠 𝐏𝐚𝐫𝐚𝐝𝐢𝐠𝐦 𝐒𝐡𝐢𝐟𝐭: 𝐖𝐡𝐞𝐧 𝐏𝐨𝐩𝐮𝐥𝐚𝐫 𝐌𝐞𝐞𝐭𝐬 𝐎𝐩𝐭𝐢𝐦𝐚𝐥 accepted at #AISTATS2026! Work done by @AntonioPuri00, @mariasofiabuc, Anil Kumar Nelakanti, @Andrea_Bacciu, @amantrac, @fabreetseo! Paper arxiv.org/pdf/2502.12581

5/5 This offers a complementary perspective on mechanistic interpretability: from finding circuits to shaping and composing them during learning. Looking forward to discussions at #ICLR! Thanks to Samuele Fonio, Simone Monaco Nicola Saccomanno and @fabreetseo !

1/5 Composable Sparse Subnetworks via the Maximum-Entropy Principle will be presented at #ICLR2026 🎉 👉 openreview.net/forum?id=IHwx5… It’s about circuits, sparsity, and compositionality, with a training-time loss that shapes how circuits form. Congrats @frances55037016 !

4/5 Crucially, these circuits are composable. We show they can be recombined into generalist models, behaving like modular building blocks rather than isolated explanations.

3/5 We introduce a training-time objective that steers circuit formation toward specific concepts. After training, we use pruning to extract class-specific circuits: functional subnetworks specialized for a subset of classes.

2/5 Most mechanistic interpretability work finds circuits post hoc, after training a model. We ask a different question: can we control which concepts shape circuits during training?

Work done by: Andrea Drago, @mariasofiabuc , @frances55037016 , Marius Michetti, @_FedeSiciliano_ , @fabreetseo and Luca Becchetti.

Paper accepted at #AISTATS2026!! 𝐑𝐚𝐧𝐤 𝐋𝐢𝐟𝐭𝐢𝐧𝐠 𝐚𝐧𝐝 𝐑𝐚𝐧𝐝𝐨𝐦 𝐍𝐨𝐧-𝐋𝐢𝐧𝐞𝐚𝐫 𝐌𝐚𝐩𝐬: once the size of a one-hidden-layer neural network exceeds a threshold, the hidden representations are linearly independent with high probability. #RSTLess #Sapienza

@chiaradercoli is 1st-year PhD student in the National Doctorate in AI, supervised by @pl219_Cambridge . 🔬 AI for Medicine | Digital Twins | ODEs 🧬 RNA pathways (T1D, Periodontitis) 🤖 LLMs for clinical & biomedical data Open to collaborations! feel free to reach out