CounterMoral: Editing Morals in Language Models
For AI safety researchers, this work highlights the challenge of aligning language models with human values through editing, but it is an incremental evaluation rather than a breakthrough.
CounterMoral introduces a benchmark to evaluate how well model editing techniques modify moral judgments in language models, finding that current methods are insufficient for reliably editing ethical beliefs.
Recent advancements in language model technology have significantly enhanced the ability to edit factual information. Yet, the modification of moral judgments, a crucial aspect of aligning models with human values, has garnered less attention. In this work, we introduce CounterMoral, a benchmark dataset crafted to assess how well current model editing techniques modify moral judgments across diverse ethical frameworks. We apply various editing techniques to multiple language models and evaluate their performance. Our findings contribute to the evaluation of language models designed to be ethical.