Steering Masked Discrete Diffusion Models via Discrete Denoising Posterior Predictionby Jarrid Rector-Brooks, Mohsin Hasan, Zhangzhi…
Reward-Augmented Data Enhances Direct Preference Alignment of LLMsby Shenao Zhang, Zhihan Liu, Boyi Liu, Yufeng…
Unstable Unlearning: The Hidden Risk of Concept Resurgence in Diffusion Modelsby Vinith M. Suriyakumar, Rohan…
SEAL: Safety-enhanced Aligned LLM Fine-tuning via Bilevel Data Selectionby Han Shen, Pin-Yu Chen, Payel Das,…
Degree Distribution based Spiking Graph Networks for Domain Adaptationby Yingxu Wang, Mengzhu Wang, Siwei Liu,…
Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Thinkby Sihyun Yu, Sangkyung…
Forgetting Through Transforming: Enabling Federated Unlearning via Class-Aware Representation Transformationby Qi Guo, Zhen Tian, Minghao…
TCGU: Data-centric Graph Unlearning based on Transferable Condensationby Fan Li, Xiaoyang Wang, Dawei Cheng, Wenjie…
Adaptive Guidance for Local Training in Heterogeneous Federated Learningby Jianqing Zhang, Yang Liu, Yang Hua,…
Honesty to Subterfuge: In-Context Reinforcement Learning Can Make Honest Models Reward Hackby Leo McKee-Reid, Christoph…