EALM: Introducing Multidimensional Ethical Alignment in Conversational Information Retrieval
This work addresses ethical alignment in conversational AI to prevent harmful information dissemination, though it is incremental as it builds on existing benchmarks and methods.
The paper tackles the problem of integrating ethical alignment into conversational information retrieval by introducing a workflow with an ethical judgment stage and new datasets, achieving top performance in binary and multi-label ethical judgment tasks.
Artificial intelligence (AI) technologies should adhere to human norms to better serve our society and avoid disseminating harmful or misleading information, particularly in Conversational Information Retrieval (CIR). Previous work, including approaches and datasets, has not always been successful or sufficiently robust in taking human norms into consideration. To this end, we introduce a workflow that integrates ethical alignment, with an initial ethical judgment stage for efficient data screening. To address the need for ethical judgment in CIR, we present the QA-ETHICS dataset, adapted from the ETHICS benchmark, which serves as an evaluation tool by unifying scenarios and label meanings. However, each scenario only considers one ethical concept. Therefore, we introduce the MP-ETHICS dataset to evaluate a scenario under multiple ethical concepts, such as justice and Deontology. In addition, we suggest a new approach that achieves top performance in both binary and multi-label ethical judgment tasks. Our research provides a practical method for introducing ethical alignment into the CIR workflow. The data and code are available at https://github.com/wanng-ide/ealm .