Year: 2026
Type: conference-paper
Abstract: Distributional alignment enables large language models (LLMs) to predict how a target population distributes its responses across answer options, rather than collapsing disagreement into a single consensus answer.However, existing LLM-based distribution prediction is often unstable and degrades under cultural and domain shift.Token score-based estimates can change with minor option wording or formatting, response sampling-based estimates are expensive and sensitive to prompts and decoding settings, and directly generated distributions are frequently miscalibrated.We propose Evi-DA 1 , an evidence-based alignment technique that improves the fidelity and robustness of LLM-based distribution estimation under domain and cultural shift.Given a target country and a multiple-choice question, Evi-DA retrieves related World Values Survey items and their answer distributions, predicts a coarse Welzel value signature for each option, and infers the country-conditioned answer distribution in a structured format.We train the LLMs using a two-stage pipeline, where reinforcement learning optimizes survey-derived rewards that encourage accurate intermediate value predictions, faithful final distributions, well-formed structured outputs, and reduced cultural bias.Across in-domain and out-ofdomain benchmarks and multiple open-source backbones, Evi-DA reduces Jensen-Shannon divergence between predicted and gold distributions relative to strong baselines, with average relative improvements of up to 44%.
Institutions: Monash University, RMIT University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166810894