@inproceedings{ijcai2026p63, title = {Explaining Jailbreaks: Structured and Interpretable Safety Assessment for Large Language Models}, author = {Dong, Sunghee and Yi, Sungwon and Bae, Kangmin and Kim, Jaeyoon}, booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, {IJCAI-26}}, publisher = {International Joint Conferences on Artificial Intelligence Organization}, editor = {Diego Calvanese}, pages = {553--561}, year = {2026}, month = {8}, note = {Main Track}, doi = {10.24963/ijcai.2026/63}, url = {https://doi.org/10.24963/ijcai.2026/63}, }