LLM-based Detection of Manipulative Political Narratives

Hugging Face Daily Papers Papers
llm detection political-narratives unsupervised-clustering prompt-filtering social-media

Summary

A computational framework combining prompt-based filtering and unsupervised clustering to identify manipulative political narrative clusters from social media posts without predefined categories.

We present a new computational framework for detecting and structuring manipulative political narratives. A task that became more important due to the shift of political discussions to social media. One of the primary challenges thereby is differentiating between manipulative political narratives and legitimate critiques. Some posts may also reframe actual events within a manipulative context. To achieve good clustering results, we filter manipulative posts beforehand using a detailed few-shot prompt that combines documented campaign narratives with legitimate criticisms to differentiate them. This prompt enables a reasoning model to assign labels, retaining only manipulative narrative posts for further processing. The remaining posts are subsequently embedded and dimensionality-reduced using UMAP, before HDBSCAN is applied to uncover narrative groups. A key advantage of this unsupervised approach is its independence from a predefined list of target categories, enabling it to uncover new narrative clusters. Finally, a reasoning model is employed to uncover the narrative behind each cluster. This approach, applied to over 1.2 million social media posts, effectively identified 41 distinct manipulative narrative clusters by integrating prompt-based filtering with unsupervised clustering.
Original Article
View Cached Full Text

Cached at: 05/15/26, 04:23 AM

Paper page - LLM-based Detection of Manipulative Political Narratives

Source: https://huggingface.co/papers/2605.14354

Abstract

A computational framework combining prompt-based filtering and unsupervised clustering identifies manipulative political narrative clusters from social media posts without requiring predefined categories.

We present a new computational framework for detecting and structuring manipulative political narratives. A task that became more important due to the shift of political discussions to social media. One of the primary challenges thereby is differentiating between manipulative political narratives and legitimate critiques. Some posts may also reframe actual events within a manipulative context. To achieve good clustering results, we filter manipulative posts beforehand using a detailedfew-shot promptthat combines documented campaign narratives with legitimate criticisms to differentiate them. This prompt enables a reasoning model to assign labels, retaining only manipulative narrative posts for further processing. The remaining posts are subsequently embedded and dimensionality-reduced usingUMAP, beforeHDBSCANis applied to uncover narrative groups. A key advantage of this unsupervised approach is its independence from a predefined list of target categories, enabling it to uncover new narrative clusters. Finally, a reasoning model is employed to uncover the narrative behind each cluster. This approach, applied to over 1.2 million social media posts, effectively identified 41 distinct manipulative narrative clusters by integratingprompt-based filteringwithunsupervised clustering.

View arXiv pageView PDFGitHub0Add to collection

Get this paper in your agent:

hf papers read 2605\.14354

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.14354 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.14354 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.14354 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Framing the Narrative: Ideological Mimicry in Large Language Models

arXiv cs.CL

The paper introduces the Poli-SHIFT dataset and evaluation framework showing that LLMs exhibit "ideological mimicry," systematically shifting political stance toward signals in user prompts, with terminology changes alone reversing model positions in 16.9% of matched comparisons across seven open-weight models.