Analyzing values about gendered language reform in LLMs' revisions
This addresses the problem of value alignment in AI language models for users concerned with gender inclusivity, though it is incremental in applying existing sociolinguistic insights to LLMs.
The study examined how LLMs revise gendered role nouns and justify their revisions, evaluating alignment with feminist and trans-inclusive language reforms in English, and found that LLMs are broadly sensitive to contextual effects similar to humans.
Within the common LLM use case of text revision, we study LLMs' revision of gendered role nouns (e.g., outdoorsperson/woman/man) and their justifications of such revisions. We evaluate their alignment with feminist and trans-inclusive language reforms for English. Drawing on insight from sociolinguistics, we further assess if LLMs are sensitive to the same contextual effects in the application of such reforms as people are, finding broad evidence of such effects. We discuss implications for value alignment.