Dataset Groups Activity Stream AdvBench dataset The dataset used for the experiments in the paper, consisting of 60 harmful instructions from the AdvBench dataset. BibTex: @dataset{Yuanpu_Cao_and_Bochuan_Cao_and_Jinghui_Chen_2024, abstract = {The dataset used for the experiments in the paper, consisting of 60 harmful instructions from the AdvBench dataset.}, author = {Yuanpu Cao and Bochuan Cao and Jinghui Chen}, doi = {10.57702/oqo7hxih}, institution = {No Organization}, keyword = {'AdvBench', 'harmful instructions', 'harmful responses', 'language models', 'question answering', 'safety alignment'}, month = {dec}, publisher = {TIB}, title = {AdvBench dataset}, url = {https://service.tib.eu/ldmservice/dataset/advbench-dataset}, year = {2024} }
