Auditing AI-Generated Product Recommendations
This matters because it reveals how biased AI recommendations can be, showing that popular chatbots like ChatGPT often favor specific products over others.
#偏见
5 stories · 0 in 7 days · 0 in 30 days
Stories, products and papers tagged "偏见", newest first.
9月17日
Linguistic Triggers of Gender and Racial Bias in Open-Weight LLMs Applied to Recruitment
This matters because it provides a concrete protocol for companies to audit their AI systems for bias before deployment, which is now required under the EU AI Act for high-risk applications like recruitment.
7月30日
Industry00:56
Chinese content模型评测中的偏见：不知道型号就觉得好，知道是前沿模型就有偏见FactoryAI的CTO聊了个扎心的现象：你不知道用啥模型时觉得都挺好，一旦知道是前沿模型就开始挑刺。模型战争原来是心理战？
7月17日
Papers11:04
Chinese content研究显示AI检测模型更易将自闭症写作误标为AI生成这篇论文发现OpenAI的GPT-2检测器对自闭症写作者有统计上显著的偏见——误判率虽低但差异存在。如果你关心AI公平性，值得看看。
7月14日
Papers12:14
Chinese contentMechanistic Interpretability Account of LLM-as-Judge Bias这篇论文用几何视角解释LLM评判偏见，发现偏见隐藏状态的低维子空间，还能预测模型在未知基准上的失败，挺有意思。