2 papers
cs.LG2026
Why Subliminal Learning Needs So Much Data: A Noisy Inverse View through Steering Vector Recovery
Luoyu Chen, Xiaoyu Ding, Weiqi Wang +4
Subliminal learning lets a student inherit a teacher's behavioral trait from semantically unrelated data, yet published demonstrations typically require tens of thousands of carrie…
cs.CR2026
Steering Beyond the Support: Adversarial Training on Unsupervised Jailbroken Activation Simulation
Luoyu Chen, Weiqi Wang, Zhiyi Tian +5
Jailbreak prompts can trigger harmful completions on aligned LLMs, In accordance, safety steering has been proposed: test-time activation interventions that steer jailbreak activat…