Defending LLMs against jailbreak attacks through representation offset detection.

Shuo Liu|Xiang Cheng|Zhenzhong Zheng|Sen Su


Anthology ID:DBLP:journals/ipm/LiuCZS26
Year:2026
Venue:Information Processing and Management
Pages:104662
URL:https://doi.org/10.1016/J.IPM.2026.104662
DOI:https://doi.org/10.1016/J.IPM.2026.104662
DBLP:journals/ipm/LiuCZS26
BibTeX:
@article{liu-2026-defending, author = {Shuo Liu and Xiang Cheng and Zhenzhong Zheng and Sen Su}, title = {{Defending LLMs against jailbreak attacks through representation offset detection}}, journal = {Information Processing and Management}, volume = {63}, number = {5}, pages = {104662}, year = {2026}, url = {https://doi.org/10.1016/J.IPM.2026.104662}, doi = {https://doi.org/10.1016/J.IPM.2026.104662} }