Technical Leader - Artificial Intelligence and Machine Learning Enthusiast - Senior Software Engineer @altenitalia
Rome, Lazio
Joined August 2011
- Tweets3.5K
- Following2.7K
- Followers1.1K
- Likes5.4K
Bruno Neri retweeted
Interpretability is the most important scientific problem of our lifetime. But it's too slow right now! I wrote some thoughts with @banburismus_ on how we might accelerate it.
(Probably a good time to share that I'll be joining @GoodfireAI post PhD. I'm over the moon about it🌝)
if any science should be revolutionised by AI, it should be interpretability. I wrote about how we can accelerate progress, and why we need a Human Genome Project for AI. coauthored with @mathildepapillo
Bruno Neri retweeted
A personal loop closes for me! Was a great honor to do the announcement today in Dublin. It’s time to make it happen
What can large-scale AI models bring to the growing complexity of particle physics?
@CERN and Mistral join forces to explore this question.
The ambition: a single Large Physics Model, trained on billions of simulated collisions.
Bruno Neri retweeted
Boltz on Times Square! Proud to see our team be part of the Disruptors 60 Class of 2026 and to see our work bringing AI-based biomolecular design to every scientist recognized. Thank you to our team, our partners and the scientists building with Boltz!
Bruno Neri retweeted
🔥Welcome 🔺Simplex Diffusion Models🔻!
Standard discrete diffusion models suffer from "information collapse" because they discard uncertainty at intermediate steps by sampling categorical tokens. (1/5)
Bruno Neri retweeted
🗓️ Join me on October 7 for one of the inaugural lectures of the Science of Intelligence Institute at @UCBerkeley !
Can we write scientific laws for intelligence? I believe that we can, that the math for it already exists, and that they apply across brains & machines.
tinyurl.com/mvrfs39w
Bruno Neri retweeted
New paper from Japan! Our work “Authorship Attribution and Aesthetic Evaluation of AI Poetry: A Case Study with Haiku” is now on arXiv and accepted as a full paper at SCIS&ISIS 2026 🎉
Work done by Livia Oddi with me, @s_scardapane and Toru Sugimoto
arxiv.org/abs/2609.15511?utm…
Bruno Neri retweeted
**Recirculation**
by @mc_mozer @ShoaibASiddiqui @savvyRL et al.
Very simple test-time intervention where the activation of a late layer is "leaked" into an early layer to steer it towards a more effective answer.
arxiv.org/abs/2608.17981
Bruno Neri retweeted
A curious property of diffusion models: when a sample is mostly noise, the model’s best guess is close to the mean.
So early on samples can be pulled toward the center of the dataset, even when no data exists there, before rebounding toward the data manifold.
Bruno Neri retweeted
🎉The results are in! We are thrilled to announce the winners of the @TAGinDS 2026 Topological Deep Learning Challenge, sponsored by @ARLQ_AI and @newtheoryai 🎉
This was our largest edition yet: 60+ submissions from 50+ participants across 15+ institutions!
🧵👇
Bruno Neri retweeted
🔴 CAREER UPDATE / I AM HIRING 🔴
After several months of being extremely annoying to everyone who knows me, I have big news: this fall I'm joining Brown University as Assistant Professor of Biostatistics, Data Science, and Computer Science!
Bruno Neri retweeted
Nice! But first take a look at link.springer.com/book/10.10… 😉
Bruno Neri retweeted
*Language Models are Injective and Hence Invertible*
by @GiorgosNik02 @tommaso_mncttn @DonatoCrisosto1 @teelinsan @EmanueleRodola
They show that decoder transformers are injective, meaning you can easily recover an input sequence from the hidden states.
arxiv.org/abs/2510.15511
Bruno Neri retweeted
the final 'G' chapter of the GDL textbook is now publicly available to read 🥹
alice welcomes you to the dynamic, exciting world of geometric graphs 🧬 -- covering neural nets for point clouds, molecules, manifolds and meshes
links & goodies in thread!
@mmbronstein @TacoCohen
Bruno Neri retweeted
Yesterday we announced the integration of Boltz with Claude Science. One thing I wanted to take a moment to highlight was the results we got from letting Claude drive the Boltz API end-to-end, so I wanted to give a bit more detail here. A thread! 🧵
Bruno Neri retweeted
Excited to share that Boltz is accessible via Claude products, including Claude Science, Anthropic’s new AI workbench for scientists!
Bruno Neri retweeted
**Stochastic activations**
by @MariaLomeli_ @DouzeMatthijs @jaseweston et al.
Simple but elegant: they randomly sample an activation function at training time for regularization, and they can also exploit it for diversity at test-time.
arxiv.org/abs/2509.22358
Bruno Neri retweeted
🧠Can geometry give us a theory of neural computation?
@claire_i_webb and I discussed how #geometry, the hidden shapes & structure inside high-dimensional neural data, can be a powerful tool to understand how the brain computes.
@longnow @berggruenInst @geometric_intel
How are scientists mapping consciousness? Could geometry hold the key? Claire Isabel Webb, director of the Future Humans program at the @berggruenInst, asks @NinaMiolane of the UCSB Geometric Intelligence Lab.
Watch the full episode here: na2.hubs.ly/H068tRX0
Bruno Neri retweeted
We’re teaming up @Palmeiras, the first football club to meaningfully build upon TacticAI: our AI system that can help simulate field scenarios and predict open play dynamics up to 8 seconds in advance. ⚽
Bruno Neri retweeted
The ocean is inherently chaotic, yet existing data-driven ocean models produce deterministic forecasts. In our new preprint, we introduce Njord, a probabilistic graph neural network for ensemble ocean forecasting.
Link: arxiv.org/abs/2605.15470
A couple highlights below 🧵
