13 citations · 35 across the 6 of their papers we have counts for
7 papers
Red-Teaming Auto Mode: Improving Blocking Classifiers Against Malign Coding Agents
Alex Remedios, Simon Storf, Fabien Roger +1
To keep coding agents from going off the rails, production systems now review each proposed action with a blocking monitor that can reject it before it runs (Auto Mode in Claude Co…
Eliciting Harmful Capabilities by Fine-Tuning On Safeguarded Outputs
Jackson Kaunismaa, Avery Griffin, John Hughes +3
Model developers implement safeguards in frontier models to prevent misuse, for example, by employing classifiers to filter dangerous outputs. In this work, we demonstrate that eve…
Why Do Some Language Models Fake Alignment While Others Don't?
Abhay Sheshadri, John Hughes, Julian Michael +4
Alignment faking in large language models presented a demonstration of Claude 3 Opus and Claude 3.5 Sonnet selectively complying with a helpful-only training objective to prevent m…
Branching Processes for QuickCheck Generators
Agustín Mista, Alejandro Russo, John Hughes
In QuickCheck (or, more generally, random testing), it is challenging to control random data generators' distributions---specially when it comes to user-defined algebraic data type…
Modelling of Autosar Libraries for Large Scale Testing
Wojciech Mostowski, Thomas Arts, John Hughes
We demonstrate a specific method and technology for model-based testing of large software projects with the QuickCheck tool using property-based specifications. Our specifications…
Beginner's Luck: A Language for Property-Based Generators
Leonidas Lampropoulos, Diane Gallois-Wong, Catalin Hritcu +3
Property-based random testing a la QuickCheck requires building efficient generators for well-distributed random data satisfying complex logical predicates, but writing these gener…