%0 Journal Article %1 vaugrante2025compromisinghonestyharmlessnesslanguage %A Vaugrante, Laurène %A Carlon, Francesca %A Menke, Maluna %A Hagendorff, Thilo %D 2025 %K %T Compromising Honesty and Harmlessness in Language Models via Deception Attacks %U https://arxiv.org/abs/2502.08301

@article{vaugrante2025compromisinghonestyharmlessnesslanguage, added-at = {2025-05-05T10:35:19.000+0200}, archiveprefix = {arXiv}, author = {Vaugrante, Laurène and Carlon, Francesca and Menke, Maluna and Hagendorff, Thilo}, biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2caefab524c34ff3c1f82ccee3a46b8f2/thilohagendorff}, eprint = {2502.08301}, interhash = {144ae9caea0c45722e8f1a4880b65c22}, intrahash = {caefab524c34ff3c1f82ccee3a46b8f2}, keywords = {}, primaryclass = {cs.CL}, timestamp = {2025-05-05T08:35:19.000+0200}, title = {Compromising Honesty and Harmlessness in Language Models via Deception Attacks}, url = {https://arxiv.org/abs/2502.08301}, year = 2025 }