Haotian Jin, Yang Li 0192, Haihui Fan, Lin Shen, Xiangfang Li, Bo Li 0063. Uncovering and Aligning Anomalous Attention Heads to Defend Against NLP Backdoor Attacks. In Sven Koenig, Chad Jenkins, Matthew E. Taylor, editors, Fortieth AAAI Conference on Artificial Intelligence, Thirty-Eighth Conference on Innovative Applications of Artificial Intelligence, Sixteenth Symposium on Educational Advances in Artificial Intelligence, AAAI 2026, Singapore, January 20-27, 2026. pages 37472-37480, AAAI Press, 2026. [doi]
Abstract is missing.