Zhebo Wang — accepted AI-conference papers

2026

FORGETMARK: STEALTHY FINGERPRINT EMBEDDING VIA TARGETED UNLEARNING IN LANGUAGE MODELS

Zhenhua Xu, Haobo Zhang, Zhebo Wang, Qichen Liu, Haitao Xu

ICASSP 2026poster

Existing invasive (backdoor) fingerprints suffer from high-perplexity triggers that are easily filtered, fixed response patterns exposed by heuristic detectors, and spurious activations on benign inputs. We introduce \textsc{ForgetMark}, a stealthy fingerprinting framework that encodes provenance vi…

Cited by 0 Source ↗PDF ↗Scholar ↗

2026

ICPO: Illocution-Calibrated Policy Optimization for Multi-Turn Conversation

Zhebo Wang, Xiaohu Mu, Zijie Zhou, Mohan Li, Wenpeng Xing

ICASSP 2026poster

Large Language Models (LLMs) in multi-turn conversations often suffer from a ``lost-in-conversation'' phenomenon, where they struggle to recover from early incorrect assumptions, particularly when users provide ambiguous initial instructions. We find that standard post-training techniques like Reinf…

Cited by 0 Source ↗PDF ↗Scholar ↗