Perception and Intelligence Lab (PINlab)

Research group at @SapienzaRoma focused on fundamental research and innovation transfer in the fields of computer vision and machine learning

[Pascal Mettes @PascalMettes]

A couple of months ago, I told you vision-language = hyperbolic. #ICLR2025 agrees and has made our latest paper an oral presentation!

To present at ICLR: Compositional Entailment Learning for Hyperbolic Vision-Language Models.

Link: https://t.co/FSBMUtqoRM

[Indro Spinelli @IndroSpinelli]

The posters are set and the @SapienzaPINlab will be ready after a very much needed cup of coffee. See you later #WACV2025

[Fabio Galasso @GalassoFab10]

"Following the Human Thread in Social Navigation" #ICLR25

has been 𝐬𝐞𝐥𝐞𝐜𝐭𝐞𝐝 𝐚𝐬 𝐬𝐩𝐨𝐭𝐥𝐢𝐠𝐡𝐭 🏅

5.1% of the submitted papers were selected for a spotlight or an oral

[Fabio Galasso @GalassoFab10]

"Compositional Entailment Learning for Hyperbolic Vision-Language Models" #ICLR25

has been 𝐬𝐞𝐥𝐞𝐜𝐭𝐞𝐝 𝐟𝐨𝐫 𝐚𝐧 𝐨𝐫𝐚𝐥 𝐩𝐫𝐞𝐬𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧 🏆

1.8% of the submitted papers were selected for an oral

[Indro Spinelli @IndroSpinelli]

AI Decodes the Brain’s "Stop" ! 🧠⚡
Our custom spatio-temporal transformer (ST-T) predicts movement inhibition in monkeys within 230ms of a "Go" signal—faster than previous methods! Another step towards BCIs and our understanding of disorders like ADHD/Parkinson’s.
👉 [1/6]

[Lamberto Ballan @lambertoballan]

#ICLR decisions are out! We have a paper on social navigation (this is an early version https://t.co/Ngai2KhDm5). Great collaboration with
@GalassoFab10 and his awesome students from @SapienzaRoma

[Fabio Galasso @GalassoFab10]

Our @SapienzaPINlab has 2 #ICLR25 papers accepted!
I'm proud of the strong team and of the fantastic collaborators.

[Alessandro Flaborea @AlessandroFlabo]

Just a quick update on our HyCoCLIP paper: accepted to the #ICLR2025 conference with a avg score of 8!

A great collaboration between @SapienzaPINlab (@GalassoFab10 , @GuidoDamely) and @PascalMettes @theAvikPal

Paper: https://t.co/RrR2wV3HTB

See the thread for more insights 👇🏻

[Leonardo Plini @LeonardoPlini]

I am very happy to share that I've received the Young Speaker Award at WIFAI 2024! 🏆
I  feel truly fortunate to be part of @SapienzaPINlab and the NA62 group at INFN-LNF.

[Alessandro Flaborea @AlessandroFlabo]

Excited to share our latest research, TI-PREGO—a follow-up to PREGO, our #CVPR2024 work! 📈

We explore how Chain of Thought and In-context learning can boost mistake detection in procedural videos.

Paper: https://t.co/75aiE3lVku

A good read before the #CVPR2025 deadline! 🚀

[Fabio Galasso @GalassoFab10]

ARIA glasses have arrived today at @SapienzaPINlab
Thanks @meta_aria and @AIatMeta for getting us onboard!

[Avik Pal @theAvikPal]

(1/6)🥳 Excited to share my latest research done as part of my MSc AI thesis! We introduced Hyperbolic Compositional CLIP (HyCoCLIP)—a novel framework that leverages the hierarchical nature of hyperbolic space for learning vision-language representations using scene compositions.

[ItalAI @_italai]

To learn how Hyperbolic Neural Networks could help managing the uncertainty in AI models, check out the papers published by our team 👇

1⃣ ICML'24 HALO: https://t.co/TpTapQSHAI
2⃣IROS'24 Hyp2Nav: https://t.co/cLXDWT5nTd

#AI #NeuralNetworks #MachineLearning #AIResearch #Tech

[Perception and Intelligence Lab (PINlab) @SapienzaPINlab]

PINlab at @TEDxPadova! Prof. @GalassoFab10 had a talk about uncertainty in AI models and how Hyperbolic Neural Networks can handle it 🔥

[ItalAI @_italai]

On Oct 16, @GalassoFab10, founder of ItalAI & head of @SapienzaPINlab, spoke at @TEDxPadova dIAloghi_Conversazioni sull'AI💡about managing uncertainty in AI & the role of Hyperbolic Neural Networks in the ongoing search for solutions.

#NeuralNetworks #MachineLearning #AI #Tech

[Alessandro Flaborea @AlessandroFlabo]

Today #IROS2024 starts and @SapienzaPINlab is present ✅

@GuidoDamely will have the oral presentation of Hyp2Nav, work done while we were visiting @PascalMettes, on the 18th.

He'll show how hyperbolic NNs improve explainability and reduce dimensionality for RL.
@GalassoFab10

[ItalAI @_italai]

On the last (but not least) day of the #ECCV2024, founder of ItalAI @GalassoFab10, together with @AlessioSampier1, @palm_alessio and @IndroSpinelli from @SapienzaPINlab, presented "Length-Aware Motion Synthesis via Latent Diffusion".

Great work 🤝

@eccvconf

[Fabio Galasso @GalassoFab10]

I look forward to discussing artificial intelligence at the upcoming TEDx in Padua.

I'll present novel insights into the beliefs and certainties of intelligent machines based on the research of @SapienzaPINlab and the collaboration with international research partners

[ItalAI @_italai]

ItalAI and @SapienzaPINlab at the #ECCV2024 in Milan today 🔥
@eccvconf

[Alessandro Flaborea @AlessandroFlabo]

Are you ready to proceed with us? 👀

Tomorrow we'll be showing our interactive demo during the morning demo session (9am-12.30)!

See you there with https://t.co/ybbIEqkAUt 😎

#ECCV2024 @eccvconf

[ItalAI @_italai]

@paolomandica and @FraLuca17 presenting their work on the first large-scale Hyperbolic vision-language model at the “Beyond Euclidean” workshop yesterday #ECCV2024.
Big thanks to those who attended and joined the dynamic discussion. It was amazing!
@eccvconf