An Assurance Case Pattern for the Interpretability of Machine Learning in Safety-Critical Systems von Ward, Francis Rhys, Habli, Ibrahim

Volltext

Argumentative Reward Learning: Reasoning About Human Preferences von Ward, Francis Rhys, Belardinelli, Francesco, Toni, Francesca

Volltext bestellen

Honesty Is the Best Policy: Defining and Mitigating AI Deception von Ward, Francis Rhys, Belardinelli, Francesco, Toni, Francesca, Everitt, Tom

Volltext bestellen

An Assurance Case Pattern for the Interpretability of Machine Learning in Safety-Critical Systems von Ward, Francis Rhys, Habli, Ibrahim

Volltext bestellen

Experiments with Detecting and Mitigating AI Deception von Sahbane, Ismail, Ward, Francis Rhys, Åslund, C Henrik

Volltext bestellen

AI Sandbagging: Language Models can Strategically Underperform on Evaluations von van der Weij, Teun, Hofstätter, Felix, Jaffe, Ollie, Brown, Samuel F, Ward, Francis Rhys

Volltext bestellen

The Reasons that Agents Act: Intention and Instrumental Goals von Ward, Francis Rhys, MacDermott, Matt, Belardinelli, Francesco, Toni, Francesca, Everitt, Tom

Volltext bestellen

Evaluating Language Model Character Traits von Ward, Francis Rhys, Yang, Zejia, Jackson, Alex, Brown, Randy, Smith, Chandler, Colverd, Grace, Thomson, Louis, Douglas, Raymond, Bartak, Patrik, Rowan, Andrew

Volltext bestellen

Argumentative Reward Learning: Reasoning About Human Preferences von Ward, Francis Rhys, Belardinelli, Francesco, Toni, Francesca

Volltext

Honesty Is the Best Policy: Defining and Mitigating AI Deception von Ward, Francis Rhys, Belardinelli, Francesco, Toni, Francesca, Everitt, Tom

Volltext

Experiments with Detecting and Mitigating AI Deception von Ismail Sahbane, Ward, Francis Rhys, Åslund, C Henrik

Volltext

AI Sandbagging: Language Models can Strategically Underperform on Evaluations von Teun van der Weij, Hofstätter, Felix, Jaffe, Ollie, Brown, Samuel F, Ward, Francis Rhys

Volltext

The Reasons that Agents Act: Intention and Instrumental Goals von Ward, Francis Rhys, MacDermott, Matt, Belardinelli, Francesco, Toni, Francesca, Everitt, Tom

Volltext