StackOne

AI & ML interests

None defined yet.

Recent Activity

hisku  submitted a paper 1 day ago

Train the Model, Not the Reader: Decodability Supervision for Verifiable Activation Explanations

hisku  authored a paper 12 months ago

When Good Sounds Go Adversarial: Jailbreaking Audio-Language Models with Benign Inputs

hisku  authored a paper 12 months ago

Eliciting and Analyzing Emergent Misalignment in State-of-the-Art Large Language Models

View all activity

Team members 4