19 papers
UniNDM: A Unified Noise-driven Detection and Mitigation Framework Against Sexual Content in Text-to-Image Generation
Yao Huang, Yitong Sun, Huanran Chen +8
Despite the impressive generative capabilities of text-to-image diffusion models, they remain vulnerable to implicit sexual prompts, where subtle cues disguised as benign terms or…
Nexus: Same Pretraining Loss, Better Downstream Generalization via Common Minima
Huanran Chen, Huaqing Zhang, Xiao Li +3
The foundational capabilities of large language models are acquired during pretraining on internet-scale, highly heterogeneous data mixtures. In this work, we investigate an intere…
Unveiling the Basin-Like Loss Landscape in Large Language Models
Huanran Chen, Yinpeng Dong, Zeming Wei +4
We discover the emergence of \textit{basins} in the loss landscape of large language models. As model scale increases, LLMs become progressively more resilient to random perturbati…
Towards Safe Reasoning in Large Reasoning Models via Corrective Intervention
Yichi Zhang, Yue Ding, Jingwen Yang +7
Although Large Reasoning Models (LRMs) have progressed in solving complex problems, their chain-of-thought (CoT) reasoning often contains harmful content that can persist even when…
Reasoning as State Transition: A Representational Analysis of Reasoning Evolution in Large Language Models
Siyuan Zhang, Jialian Li, Yichi Zhang +3
Large Language Models have achieved remarkable performance on reasoning tasks, motivating research into how this ability evolves during training. Prior work has primarily analyzed…
Scaling Laws for Black box Adversarial Attacks
Chuan Liu, Huanran Chen, Yichi Zhang +2
Adversarial examples exhibit cross-model transferability, enabling threatening black-box attacks on commercial models. Model ensembling, which attacks multiple surrogate models, is…