Author

Sripad Karne

3 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Oct 2026

What the Guard Misses, the Robot Executes: Implied Harm in VLA Instructions

Vision-language-action models (VLAs) act on instructions without being able to refuse, so screening harmful requests falls to monitors. We test whether these monitors catch ordinary robot tasks requested for harmful reasons, holding the task fixed while varying only how explicitly the intent is stated. $\pi_{0.5}$ comp...

Sripad Karne, Arjun Balaji · 0 citations

Preprint Sep 2026

Safety Monitors Mostly Catch What the Model Already Refuses

This work measures recall on exactly those prompts, defined by sampling the target model and judging its responses, defined by sampling the target model and judging its responses.

Sripad Karne · 1 citation

#natural language process... Preprint Sep 2026

Recall Is Not Protection: Evaluating Safety Monitors Against Model Compliance

Safety monitors screen prompts sent to deployed language models, flagging harmful requests so they are never answered. They are evaluated by recall against harmfulness labels, but a catch only prevents harm if the model would otherwise have complied. We measure the difference directly: we sample repeated responses from...

Sripad Karne · 0 citations