5 papers
Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges
Xiaohua Wang, Muzhao Tian, Yuqi Zeng +20
Reinforcement Learning from Human Feedback (RLHF) and related alignment paradigms have become central to steering large language models (LLMs) and multimodal large language models…
SATER: A Self-Aware and Token-Efficient Approach to Routing and Cascading
Yuanzhe Shen, Yide Liu, Zisu Huang +3
Large language models (LLMs) demonstrate remarkable performance across diverse tasks, yet their effectiveness frequently depends on costly commercial APIs or cloud services. Model…
IntentionReasoner: Facilitating Adaptive LLM Safeguards through Intent Reasoning and Selective Query Refinement
Yuanzhe Shen, Zisu Huang, Zhengkang Guo +5
The rapid advancement of large language models (LLMs) has driven their adoption across diverse domains, yet their ability to generate harmful content poses significant safety chall…
Improving Continual Pre-training Through Seamless Data Packing
Ruicheng Yin, Xuan Gao, Changze Lv +3
Continual pre-training has demonstrated significant potential in enhancing model performance, particularly in domain-specific scenarios. The most common approach for packing data b…
Explainable Synthetic Image Detection through Diffusion Timestep Ensembling
Yixin Wu, Feiran Zhang, Tianyuan Shi +7
Recent advances in diffusion models have enabled the creation of deceptively real images, posing significant security risks when misused. In this study, we empirically show that di…