Enhancing Pre-training Data Detection in LLMs Through Discriminative and Symmetric Prefix Selection

Kai Sun, Yuxin Lin, Bo Dong 0001, Jingyao Zhang, Bin Shi. Enhancing Pre-training Data Detection in LLMs Through Discriminative and Symmetric Prefix Selection. 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 33100-33107, AAAI Press, 2026. [doi]

Authors

Kai Sun

This author has not been identified. Look up 'Kai Sun' in Google

Yuxin Lin

This author has not been identified. Look up 'Yuxin Lin' in Google

Bo Dong 0001

This author has not been identified. Look up 'Bo Dong 0001' in Google

Jingyao Zhang

This author has not been identified. Look up 'Jingyao Zhang' in Google

Bin Shi

This author has not been identified. Look up 'Bin Shi' in Google