Skip to content
View mihail-gribov's full-sized avatar
1 follower · 2 following

Achievements

Achievement: Pull SharkAchievement: YOLO

Achievements

Achievement: Pull SharkAchievement: YOLO

Block or report mihail-gribov

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
mihail-gribov/README.md

Three things decide whether an AI system can be trusted: what the model is doing inside, what data stands behind it, and what the harness around it will not let through. I work on all three, in that order, because each one is what makes the next possible.

Datasets on Hugging Face · findings at subsemantic.com · notes on LinkedIn and X

Pinned Loading

  1. quadrat-ipi-eval quadrat-ipi-eval Public

    Evaluation harness for indirect prompt-injection detectors: recall by injection type, measured at 0.1% and 1% false positives

    Python