TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process Rewards - researchr publication authors

researchr

You are not signed in
Sign in
Sign up

Xiqiao Xiong, Ouxiang Li, Zhuo Liu, Moxin Li, Wentao Shi 0002, Fengbin Zhu, Qifan Wang 0001, Fuli Feng. TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process Rewards. In Maria Liakata, Viviane P. Moreira, Jiajun Zhang 0001, David Jurgens, editors, Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2026, San Diego, California, United States, July 2-7, 2026. pages 48086-48109, Association for Computational Linguistics, 2026. [doi]

This author has not been identified. Look up 'Xiqiao Xiong' in GoogleThis author has not been identified. Look up 'Ouxiang Li' in GoogleThis author has not been identified. Look up 'Zhuo Liu' in GoogleThis author has not been identified. Look up 'Moxin Li' in GoogleThis author has not been identified. Look up 'Wentao Shi 0002' in GoogleThis author has not been identified. Look up 'Fengbin Zhu' in GoogleThis author has not been identified. Look up 'Qifan Wang 0001' in GoogleThis author has not been identified. Look up 'Fuli Feng' in Google

runs on WebDSL