Natural Language Decompositions of Implicit Content Enable Better Text Representations
This work addresses the challenge of better understanding human interpretations in NLP, particularly for social science applications, though it is incremental in applying existing models to new inference tasks.
The paper tackles the problem of interpreting text by modeling implicit content, using a large language model to generate and validate propositions related to observed text. The method improves performance in tasks like argument similarity assessment, opinion data analysis, and legislative behavior modeling, with results showing enhanced text representations.
When people interpret text, they rely on inferences that go beyond the observed language itself. Inspired by this observation, we introduce a method for the analysis of text that takes implicitly communicated content explicitly into account. We use a large language model to produce sets of propositions that are inferentially related to the text that has been observed, then validate the plausibility of the generated content via human judgments. Incorporating these explicit representations of implicit content proves useful in multiple problem settings that involve the human interpretation of utterances: assessing the similarity of arguments, making sense of a body of opinion data, and modeling legislative behavior. Our results suggest that modeling the meanings behind observed language, rather than the literal text alone, is a valuable direction for NLP and particularly its applications to social science.