LLM-based Few-Shot Early Rumor Detection with Imitation Agent

2025-12-20

AI Summary

Early rumor detection (EARD) under data-scarce conditions—i.e., accurately identifying the earliest discriminative timestamp in a rumor propagation sequence with minimal labeled samples—remains a critical challenge. Method: We propose a lightweight collaborative framework that decouples temporal decision-making from semantic judgment: (i) a trainable autonomous imitation agent locates the optimal detection timestamp, and (ii) large language models (e.g., LLaMA, ChatGLM) are frozen—requiring zero parameter updates—and perform rumor verification solely via prompt-driven inference. Contribution/Results: To our knowledge, this is the first method enabling few-shot EARD without costly LLM fine-tuning or inference overhead. It significantly improves both detection accuracy and earliness, consistently outperforming state-of-the-art approaches across four real-world datasets.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Early Rumor Detection (EARD) aims to identify the earliest point at which a claim can be accurately classified based on a sequence of social media posts. This is especially challenging in data-scarce settings. While Large Language Models (LLMs) perform well in few-shot NLP tasks, they are not well-suited for time-series data and are computationally expensive for both training and inference. In this work, we propose a novel EARD framework that combines an autonomous agent and an LLM-based detection model, where the agent acts as a reliable decision-maker for extit{early time point determination}, while the LLM serves as a powerful extit{rumor detector}. This approach offers the first solution for few-shot EARD, necessitating only the training of a lightweight agent and allowing the LLM to remain training-free. Extensive experiments on four real-world datasets show our approach boosts performance across LLMs and surpasses existing EARD methods in accuracy and earliness.

Problem

• Detect early rumors from social media posts
• Address data scarcity in few-shot learning scenarios
• Reduce computational costs of LLMs for time-series data

Innovation

• Agent determines early detection time points
• LLM serves as training-free rumor detector
• Lightweight agent trained for few-shot scenarios