The framework, Recursive Environment Scaling (RES), transforms existing benchmark environments while preserving their objectives and completion requirements and identifies recursive inheritance and dependency composition as important contributors to increased difficulty.
Lin Jing, Zheng-Bo Jiao, Yi-Chen Xiang et al.· 0 citations
Emotional expression can influence the safety decisions of large language models (LLMs), offering a potential avenue for improving safety alignment. Existing studies have mainly focused on how emotional expressions facilitate attacks under harmful requests, while overlooking their effects on benign requests. We find th...
Shu-Yi Miao, Yao-Jin Ma, Chen-Hang Cui et al.· 0 citations
This work proposes aligning cross-lingual thoughts and responses (ACTR), a framework that improves multilingual safety alignment by strengthening the use of existing safety reasoning, and devise neuron-selective consistency optimization (NSCO), which uses a frozen judge model to reward agreement between the safety cate...