Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a framework for conditional hypothesis generation that incorporates researcher-specified covariates to enhance the interpretability of language variations in LLM-based text analysis. By addressing challenges such as stratum imbalance and sign reversal, the authors propose econometrics-inspired methods that improve hypothesis discovery within relevant subgroups. Experimental results demonstrate that their approach outperforms traditional global baselines, yielding more meaningful hypotheses tailored to specific contexts.
Ignoring covariates in LLM-based hypothesis generation can lead to misleading conclusions, but a new framework ensures that generated hypotheses are both relevant and interpretable.
A core goal of computational social science is to discover interpretable differences in how language varies across outcomes of interest, such as political affiliation or instructional quality. Recent LLM-based hypothesis generation methods describe such differences in natural language, but select for globally discriminative patterns without accounting for covariates that shape the data based on researchers'domain knowledge. When covariates are ignored, selected patterns can reflect confounds rather than differences of substantive interest. We introduce conditional hypothesis generation, a framework that incorporates researcher-specified covariates to steer hypothesis discovery toward differences that hold within relevant subgroups. Two challenges arise: the target subgroup may be underrepresented (stratum imbalance), and the direction of a difference may reverse across subgroups (sign reversal). We propose two econometrics-inspired methods: one introduces feature--covariate interactions to detect sign reversals, and the other applies within-stratum demeaning and inverse-frequency reweighting to equalize underrepresented strata. Synthetic experiments show each method outperforms global baselines in its targeted setting, and expert evaluation on two real-world datasets confirms that covariate-aware generation surfaces more useful hypotheses within relevant subgroups.