Lorenzo Basile

Postdoc @AreaSciencePark

[Lorenzo Basile @lorebasile]

Just landed in San Diego to present "Head Pursuit: Probing Attention Specialization in Multimodal Transformers", our spotlight paper @NeurIPSConf!

Don't miss poster #1013 on Wednesday, Dec 3 at 11AM.

[1/6]

[Valentino Maiorca @ValeMaiorca]

🎨Activation steering can reliably push a text-to-image generator toward a visual concept, but at a cost: each concept needs its own estimation.

⚡HyperTransport (HT) predicts the intervention directly, matching per-concept SOTA at 3–4 orders of magnitude less cost.

[1/6]

[Valentino Maiorca @ValeMaiorca]

Steering generative models with LinEAS is:
– parameter-efficient
– fine-grained
– data-efficient
– robust across models and modalities

Come check it out at #NeurIPS2025!

🗓️ Dec 5 • 4:30–7:30 PM PST

📍 Exhibit Hall C,D,E #3604

[Alberto Cazzaniga @albecazzaniga]

Come find us at

NeurIPS Poster Session 6 #1114 4PM - 7PM

to hear how multimodal models develope localized image-text communication in their internal representations.

#NeurIPS2025
@alexserra1998 @francescortu @lorebasile @DiegoDoimo

[Lorenzo Basile @lorebasile]

Amazing collaboration with @ValeMaiorca, @DiegoDoimo, @FrancescoLocat8 and @albecazzaniga.

📘 Paper: https://t.co/v3gOSVoJ2U

💻 Code: https://t.co/2dElyKxuVS

@AreaSciencePark @SapienzaRoma @ISTAustria

[6/6]

[Lorenzo Basile @lorebasile]

Just landed in San Diego to present "Head Pursuit: Probing Attention Specialization in Multimodal Transformers", our spotlight paper @NeurIPSConf!

Don't miss poster #1013 on Wednesday, Dec 3 at 11AM.

[1/6]

[Lorenzo Basile @lorebasile]

🌈 Property enhancement/inhibition

We can control image captioning with head-level interventions.

Rescaling heads specialized on colors or sentiments we can promote or remove related words, without finetuning.

[5/6]

[Area Science Park @AreaSciencePark]

🤖Seminar Series
Next week, Giorgos Nikolaou and Tommaso Mencattini (@EPFL) will present "Language Models are Injective and Hence Invertible". @GiorgosNik02 @tommaso_mncttn
📅Nov 27, 11 CET (Online)
👉Register to attend: https://t.co/MhXgYqI5bD

[Alberto Cazzaniga @albecazzaniga]

Excited to share that 2/2 papers from our Lab (LADE) were accepted to #NeurIPS2025 (one spotlight 🎉)

Great work from all the students and collaborators involved!

@AreaSciencePark @aleserra1998 @lorebasile @francescortu @lucrevaleriani @DiegoDoimo @ansuin @FrancescoLocat8

[francesco ortu @francescortu]

Thrilled to announce that our paper is accepted at #NeurIPS 2025!! See you in San Diego! 🇺🇸

[francesco ortu @francescortu]

🚨🚨 Excited to share our latest paper, now on @arxiv!

🖼️ We studied how unified VLMs, trained to generate both text and images (e.g., @MetaAI's Chameleon), exchange information between modalities, comparing them to standard VLMs.

Deep dive:👇

[Diego @DiegoDoimo]

I just landed in Vancouver to present @NeurIPSConf the findings of our new work!

Few-shot learning and fine-tuning change the hidden layers of LLMs in a dramatically different way, even when they perform equally well on multiple-choice question-answering tasks.
🧵1/6

[Valentino Maiorca @ValeMaiorca]

✨ Meet #ResiDual, a novel perspective on the alignment of multimodal latent spaces!

Think of it as a spectral "panning for gold" along the residual stream. It improves text-image alignment by simply amplifying task-related directions! 🌌🔍

https://t.co/UuXoYBBsT5

[1/6]