We’re sharing the next major milestone in our non-invasive brain-to-text decoder research: Brain2Qwerty v2.
Building on v1, which was published today in @Nature, Brain2Qwerty v2 is the highest-performing end-to-end pipeline capable of real-time sentence decoding from raw brain signals. It advances beyond character-level performance to decoding words and semantics, enabling accuracy for overall communication.
We believe this research has the potential to make a real difference for the millions of people who suffer from brain lesions or disorders that prevent them from communicating.
🧵👇
How to verify Merkle roots on-chain on Solana?
Feels like there’s no simple reusable library — SPL compression relies on off-chain indexers to reconstruct the tree and generate proofs.
For off-chain state with only the root stored on-chain, do most teams end up implementing their custom Merkle proof verifiers? @solana@solana_devs@kirat_tw
Solana's Alpenglow consensus is live on a community cluster.
The protocol that promises sub-second finality is being stress-tested in the wild right now.
Here's how it actually works.🧵
Building an intuition of the Agentic Systems and you always come back to where it started. ReAct is the paper that gave agents their fundamental loop: Thought, Action, Observation. Chain of Thought taught models to reason. Action-only models could use tools. ReAct was the first to show that interleaving both — letting the model think, act, observe, and think again — unlocks a completely different class of capability"