1

FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction
Runqi Lin, Alasdair Paren, Suqin Yuan, Muyang Li, Philip Torr, Adel Bibi, Tongliang Liu
PDF Code
FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction
Beyond Linear Probes: Dynamic Safety Monitoring for Language Models
James Oldfield, Philip Torr, Ioannis Patras, Adel Bibi, Fazl Barez
PDF Code
Beyond Linear Probes: Dynamic Safety Monitoring for Language Models
BiasBusters: Uncovering and Mitigating Tool Selection Bias in Large Language Models
Thierry Blankenstein, Jialin Yu, Zixuan Li, Vassilis Plachouras, Sunando Sengupta, Philip Torr, Yarin Gal, Alasdair Paren, Adel Bibi
PDF Code
BiasBusters: Uncovering and Mitigating Tool Selection Bias in Large Language Models
Attacking multimodal OS agents with malicious image patches
Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software. Create your slides in Markdown - click the Slides button to check out the example.
Lukas Aichberger, Alasdair Paren, Guohao Li, Philip H.S. Torr, Yarin Gal, Adel Bibi
PDF Project Video
Attacking multimodal OS agents with malicious image patches
Measuring what Matters: Construct Validity in Large Language Model Benchmarks
Andrew M Bean, Ryan Othniel Kearns, Angelika Romanou, Franziska Sofia Hafner, Harry Mayne, Jan Batzner, Negar Foroutan, Chris Schmitz, Karolina Korgul, Hunar Batra, Oishi Deb, Emma Beharry, Cornelius Emde, Thomas Foster, Anna Gausen, María Grandury, Simeng Han, Valentin Hofmann, Lujain Ibrahim, Hazel Kim, Hannah Rose Kirk, Fangru Lin, Gabrielle Kaili-May Liu, Lennart Luettgau, Jabez Magomere, Jonathan Rystrøm, Anna Sotnikova, Yushi Yang, Yilun Zhao, Adel Bibi, Antoine Bosselut, Ronald Clark, Arman Cohan, Jakob Foerster, Yarin Gal, Scott A Hale, Inioluwa Deborah Raji, Christopher Summerfield, Philip HS Torr, Cozmin Ududec, Luc Rocher, Adam Mahdi
PDF Dataset Project
Measuring what Matters: Construct Validity in Large Language Model Benchmarks
On the Coexistence and Ensembling of Watermarks
Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software. Create your slides in Markdown - click the Slides button to check out the example.
Aleksandar Petrov, Shruti Agarwal, Philip H.S. Torr, Adel Bibi, John Collomosse
PDF
On the Coexistence and Ensembling of Watermarks
Rethinking Safety in LLM Fine-tuning: An Optimization Perspective
Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software. Create your slides in Markdown - click the Slides button to check out the example.
Minseon Kim, Jin Myung Kwak, Lama Alssum, Bernard Ghanem, Philip Torr, David Krueger, Fazl Barez, Adel Bibi
PDF
Rethinking Safety in LLM Fine-tuning: An Optimization Perspective
Mixture of Experts Made Intrinsically Interpretable
Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software. Create your slides in Markdown - click the Slides button to check out the example.
Xingyi Yang, Constantin Venhoff, Ashkan Khakzar, Christian Schroeder de Witt, Puneet K Dokania, Adel Bibi, Philip Torr
PDF
Mixture of Experts Made Intrinsically Interpretable
Bi-Factorial Preference Optimization: Balancing Safety-Helpfulness in Language Models
Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software. Create your slides in Markdown - click the Slides button to check out the example.
Wenxuan Zhang, Philip H.S. Torr, Mohamed Elhoseiny, Adel Bibi
PDF Code
Bi-Factorial Preference Optimization: Balancing Safety-Helpfulness in Language Models
Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language Models
Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software. Create your slides in Markdown - click the Slides button to check out the example.
Francisco Eiras, Aleksandar Petrov, Philip H.S. Torr, M. Pawan Kumar, Adel Bibi
PDF
Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language Models