4 papers
Structured Confidence Perturbation via Fisher Geometry for Adversarial Training
Chenlong Wei, Hanling Wang, Chong Zhang +3
Adversarial training is often plagued by "pseudo-attacks", perturbations that significantly increase the loss value but fail to truly alter the model's confidence structure. To add…
SonicWorld: Source-Centric Latent Dynamics for Spatial Audio Reasoning
Chenlong Wei, Hanling Wang, Mingjin Lou +6
Dynamic spatial audio reasoning requires a model to associate changing acoustic observations with the correct source and recover its spatial state at a queried time or interval, a…
SyRHM: Symbolic-Language-Enhanced Reasoning with Associative Retrieval for Zero-shot Harmful Meme Detection
Hanling Wang, Chenlong Wei, Yingjuan Li +4
Detecting harmful memes is critical for maintaining safe online communities. However, harmful intent is often implicit, arising from visual-textual incongruity and cultural stereot…
ForeSight: Enhancing Risk Monitoring via Early Safety Signal Distillation
Hanling Wang, Chenlong Wei, Ling Xu +6
As large language models (LLMs) are increasingly deployed, the generation of harmful content has become a critical safety concern. Existing safeguards operate at the input, output,…