Kemou Li

Forbidden capabilities can still be learned even when suppressed in current outputs, as internal gradients enable future acquisition, but gradient sealing effectively blocks these pathways.

Latent LLM personas persist in weights post-training, readily resurfacing with specific prompts, and standard unlearning methods are insufficient to remove them without performance degradation.

CASCADE redefines backdoor detection in multimodal learning by achieving near-perfect accuracy while effectively managing the ambiguity of poisoned samples.

Even state-of-the-art AI-generated image detectors struggle when images are cropped, resized, or compressed, revealing a critical gap in real-world robustness.

Forget fine-tuning or API calls: $k$NNProxy aligns proxy LLMs for generated text detection using a training-free $k$-NN retrieval method, achieving robust performance even when the source LLM is a black box.