VQV Signal
According to a post on X by @ilblackdragon, Delta League is an open contest from a project led by NEAR co-founder Illia Polosukhin. Anyone can submit changes to model weights, and submissions are scored on a hidden test set. The post links the launch to this week's discussion about LLMs editing weights directly, and says the project plans to add architecture changes and custom domain evaluations.
According to a post on X by @ilblackdragon, Delta League is an open contest from a project led by NEAR co-founder Illia Polosukhin. Anyone can submit changes to model weights, and submissions are scored on a hidden test set. The post links the launch to this week's discussion about LLMs editing weights directly, and says the project plans to add architecture changes and custom domain evaluations.
AI-assisted summary based on the listed source.
NEAR co-founder Illia Polosukhin's project lets anyone submit changes to model weights that are scored on a hidden test set, following this week's buzz about LLMs editing weights directly. It plans to add architecture changes and custom domain evaluations. Why it matters: It turns a viral research idea into an...
The post presents the project as a way to turn a viral research idea into an open, verifiable benchmark. Builders working on small open models could use it to test weight changes against a shared hidden test set. Watch whether the planned architecture changes and custom domain evaluations are added.
VQV organizes public signals from inspectable sources. It does not independently verify the underlying report.
Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.
VQV surfaced this signal because it is recent, relevant to AI Search, connected to Scobleizer News Pulse (X).
No login, cookies, social SDKs, or automatic posting.
