A Consensus-Driven Multi-LLM Pipeline for Missing-Person Investigations

TL;DR

This paper introduces Guardian, an end-to-end multi-LLM pipeline for missing-person investigations that leverages consensus among models and fine-tuning to improve information extraction and decision support within critical early search phases.

Contribution

It presents a novel multi-LLM pipeline with consensus-driven output reconciliation and QLoRA-based fine-tuning tailored for missing-person search operations.

Findings

Effective consensus mechanism improves extraction accuracy

Fine-tuning enhances model reliability and consistency

Supports early search planning with structured information extraction

Abstract

The first 72 hours of a missing-person investigation are critical for successful recovery. Guardian is an end-to-end system designed to support missing-child investigation and early search planning. This paper presents the Guardian LLM Pipeline, a multi-model system in which LLMs are used for intelligent information extraction and processing related to missing-person search operations. The pipeline coordinates end-to-end execution across task-specialized LLM models and invokes a consensus LLM engine that compares multiple model outputs and resolves disagreements. The pipeline is further strengthened by QLoRA-based fine-tuning, using curated datasets. The presented design aligns with prior work on weak supervision and LLM-assisted annotation, emphasizing conservative, auditable use of LLMs as structured extractors and labelers rather than unconstrained end-to-end decision makers.