Pal, Soumyadeep

3 publications

ICLR 2026 Unlearning Isn't Invisible: Detecting Unlearning Traces in LLMs from Model Outputs Yiwei Chen, Soumyadeep Pal, Yimeng Zhang, Qing Qu, Sijia Liu

ICML 2025 Invariance Makes LLM Unlearning Resilient Even to Unanticipated Downstream Fine-Tuning Changsheng Wang, Yihua Zhang, Jinghan Jia, Parikshit Ram, Dennis Wei, Yuguang Yao, Soumyadeep Pal, Nathalie Baracaldo, Sijia Liu

ICLR 2024 Backdoor Secrets Unveiled: Identifying Backdoor Data with Optimized Scaled Prediction Consistency Soumyadeep Pal, Yuguang Yao, Ren Wang, Bingquan Shen, Sijia Liu