Twitxy Saved

Adi Mashiach, MD

Verified account This account is verified. Learn more

@AdiMashiach

Co-founder @OutputLabsAI. Building Mechanistic Reasoning, AI that learns how and why complex systems work. 3rd-time founder (NASDAQ IPO + acquisition).
New York, NY outputlabs.ai Joined January 2016
373 Following 296 Followers 83 Posts

Latest posts 5

Adi Mashiach, MD Verified account @AdiMashiach
Thanks to everyone who reached out yesterday. If you only saw the hit rate, the chart below is the part worth a second look. These abilities showed up when we scaled the model to 1.6B, without us ever training for them, and I think it's only the start of what Bonbon can learn. More at #NeurIPS2026 .
Output Labs Verified account @OutputLabsAI 2d
Yesterday we announced Bonbon, and a lot of the conversation focused on the 61.5% hit rate. We understand why, but for us the other exciting result is the one in this chart.

We trained Bonbon on protein and molecule sequences alone, without any structures or labels. At smaller sizes, its scores on these tasks barely changed. When we scaled it to 1.6B parameters, they jumped, even though the training loss gave no sign that this was coming. It makes us think there is a lot more in these sequences still to find.

Adi will present Bonbon as an Oral at #NeurIPS2026 . If you missed the blog, you can find it in the replies.
View quoted post on X
1 replies 0 reposts 16 likes 661 views View on X
Adi Mashiach, MD Verified account @AdiMashiach
Today, improving one property of a molecule can undermine another, sending a drug discovery program through another round of redesign and testing. Shared interaction representations create a path to considering those requirements together from the outset, making more ambitious combinations of properties practical to pursue.
Pranam Chatterjee Verified account @pranamanam 3d
@AdiMashiach and the Output Bio team have single-handedly made me believe that small molecule design is possible. 💊 Bonbon is an unbelievably powerful multimodal model in terms of its representation space (we've used it for design)! A fully deserved #NeurIPS2026 Oral! 👏
View quoted post on X
0 replies 1 reposts 14 likes 580 views View on X