LLM-generated political discourse across crises is fluent yet caricatured: more negative, less emotionally varied, more structurally regular, and lexically abstract than observed online populations.
A public dataset tracking social media discourse about the 2024 us presidential election on twitter/x
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
SMDT is a Python toolkit that standardizes social network data into a generic schema of Communities, Accounts, Posts, Actions, and Entities, with built-in anonymization and enrichment modules to support reproducible multi-platform research.
Large-scale analysis of election tweets finds highest toxicity intensity in identity issues, harassment as the dominant harm type, partisan posts more toxic than neutral with issue-varying asymmetries, and toxic content driven by high-arousal negative emotions plus context-shaped moral foundations.
citing papers explorer
-
The Algorithmic Caricature: Auditing LLM-Generated Political Discourse Across Crisis Events
LLM-generated political discourse across crises is fluent yet caricatured: more negative, less emotionally varied, more structurally regular, and lexically abstract than observed online populations.
-
Social Media Data Toolkit: Standardization and Anonymization of Social Network Datasets
SMDT is a Python toolkit that standardizes social network data into a generic schema of Communities, Accounts, Posts, Actions, and Entities, with built-in anonymization and enrichment modules to support reproducible multi-platform research.
-
Mapping Election Toxicity on Social Media across Issue, Ideology, and Psychosocial Dimensions
Large-scale analysis of election tweets finds highest toxicity intensity in identity issues, harassment as the dominant harm type, partisan posts more toxic than neutral with issue-varying asymmetries, and toxic content driven by high-arousal negative emotions plus context-shaped moral foundations.