5 papers
Visual prompt engineering for video models
Robert Geirhos, Yuxuan Li, Thaddäus Wiedemer +7
In the age of foundation models, a model is only as good as its prompt. For this reason, prompt engineering has become an essential technique for improving language model performan…
The ACUTE Protocol: Operationalizing Language Model Activations for Better Calibration, Utility, and Trust
Nishant Subramani, Palash Goyal, Yiwen Song +4
As language models improve and become increasingly deployed to solve a variety of tasks, trustworthiness becomes essential. Calibration is a good proxy for trust: well-calibrated c…
Hiding in Plain Text: Detecting Concealed Jailbreaks via Activation Disentanglement
Amirhossein Farzam, Majid Behabahani, Mani Malek +2
Large language models (LLMs) remain vulnerable to jailbreak prompts that are fluent and semantically coherent, and therefore difficult to detect with standard heuristics. A particu…
Interpreting and Controlling Model Behavior via Constitutions for Atomic Concept Edits
Neha Kalibhat, Zi Wang, Prasoon Bajpai +4
We introduce a black-box interpretability framework that learns a verifiable constitution: a natural language summary of how changes to a prompt affect a model's specific behavior,…
ShieldGemma 2: Robust and Tractable Image Content Moderation
Wenjun Zeng, Dana Kurniawan, Ryan Mullins +14
We introduce ShieldGemma 2, a 4B parameter image content moderation model built on Gemma 3. This model provides robust safety risk predictions across the following key harm categor…