Alessandro Serra
alexpietroserra.bsky.social
Alessandro Serra
@alexpietroserra.bsky.social
🎉 Update: Our paper has been ACCEPTED to NeurIPS 2025! See you in San Diego!
🚨 🚨 Excited to share our latest paper, now on #arXiv!

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

📄 Paper: arxiv.org/abs/2412.06646

Deep dive: 👇
September 24, 2025 at 7:40 AM
Pandas and multimodality, what else?
Excited to share our latest work on multimodality and interpretability in VLMs! Huge thanks to my incredible coauthors for an inspiring collaboration.
🚨 🚨 Excited to share our latest paper, now on #arXiv!

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

📄 Paper: arxiv.org/abs/2412.06646

Deep dive: 👇
December 11, 2024 at 9:04 PM
Thrilled to collaborate with an amazing team on this paper! I'm very grateful for the opportunity and proud of our work together. Check out the full paper here: arxiv.org/abs/2409.03662
December 11, 2024 at 8:55 PM