SCBO: Semantically Coherent Batching and Ordering for LLM-Based Social Surveys

2026-09-28

AI Summary

This study addresses the efficiency and accuracy bottlenecks in large language model (LLM)-based survey simulation caused by redundant context encoding, limited referencing, and the absence of answer dependencies. We propose a training-free multi-question joint prediction framework that introduces a novel semantically coherent batching mechanism. This mechanism enables efficient context sharing through LLM semantic representation extraction, template noise filtering, and centroid completion. Furthermore, it enhances reasoning capabilities by integrating target-specific retrieval with an easy-to-hard intelligent ordering strategy. Experiments on four large-scale datasets demonstrate that the proposed method significantly reduces token consumption and inference latency while effectively improving prediction accuracy.

Abstract

Large Language Models (LLMs) offer a scalable way to simulate survey respondents using demographic profiles and observed reference responses. However, the conventional approach of predicting one question per prompt repeatedly encodes the same context, limits each target to a narrow set of reference responses, and prevents later predictions from using information in earlier answers. Predicting multiple questions in one prompt can reduce these costs, share a broader pool of references, and let later predictions build on earlier ones. This requires forming coherent batches, selecting shared references, and ordering questions and references effectively. We propose Semantically Coherent Batching and Ordering (SCBO), a training-free framework that addresses these challenges. SCBO first uses an LLM to extract compact semantic representations from survey items and filter out template noise. It then groups related questions into batches and builds a shared reference bank using target-specific retrieval and centroid-based completion. Finally, it orders target questions from easy to hard and arranges references according to their semantic alignment with those questions. Experiments on four large-scale survey datasets and four LLMs show that SCBO substantially reduces token consumption and inference time while generally improving prediction accuracy over a non-batched baseline. Code is available at https://anonymous.4open.science/r/SCBO-41D8.

Problem

• Large Language Models
• Social Surveys
• Batching
• Ordering
• Token Efficiency

Innovation

• Semantically Coherent Batching
• Training-free Framework
• Survey Simulation
• Reference Ordering
• Large Language Models