arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.00871cs.CL
收藏

利用上下文学习提升对话安全性

Using In-Context Learning to Improve Dialogue Safety

Mila(米拉研究所);McGill University(麦吉尔大学);Amazon Alexa AI(亚马逊Alexa人工智能部门);Carnegie Mellon University(卡内基梅隆大学);Facebook(脸书)

Nicholas Meade, Spandana Gella, Devamanyu Hazarika, Prakhar Gupta, Di Jin, Siva Reddy, Yang Liu, Dilek Hakkani-Tür

AI总结:

针对大型对话模型易被诱导生成有害内容的问题,提出一种基于检索和上下文学习的方法,通过检索类似上下文的安全回复示例引导模型生成安全内容,无需训练即可媲美强基线,并提出重排序程序进一步提升安全性。

摘要:

尽管基于大型神经网络的对话模型已成为越来越熟练的对话智能体,但近期研究强调了这些系统的安全问题。例如,这些系统可能被诱导生成有害内容,这通常会延续社会偏见或刻板印象。我们研究了一种基于检索的方法,用于减少聊天机器人回复中的偏见和有害性。该方法利用上下文学习引导模型生成更安全的内容。具体而言,为了针对不安全的对话上下文生成回复,我们检索针对类似对话上下文的安全回复示例。我们发现,我们的方法无需训练即可与强基线模型表现相当。例如,通过自动评估,我们发现最佳的微调基线模型在DiaSafety数据集的不安全对话上下文中生成安全回复的比例仅比我们的方法高4.04%。最后,我们还提出了一种重排序程序,可以进一步提升回复的安全性。

英文摘要:

While large neural-based conversational models have become increasingly proficient dialogue agents, recent work has highlighted safety issues with these systems. For example, these systems can be goaded into generating toxic content, which often perpetuates social biases or stereotypes. We investigate a retrieval-based method for reducing bias and toxicity in responses from chatbots. It uses in-context learning to steer a model towards safer generations. Concretely, to generate a response to an unsafe dialogue context, we retrieve demonstrations of safe responses to similar dialogue contexts. We find our method performs competitively with strong baselines without requiring training. For instance, using automatic evaluation, we find our best fine-tuned baseline only generates safe responses to unsafe dialogue contexts from DiaSafety 4.04% more than our approach. Finally, we also propose a re-ranking procedure which can further improve response safeness.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑