Researchers have developed TENSOR, a novel unsupervised approach for detecting users engaged in Information Operations (IO) on social media. TENSOR combines behavioral and language patterns, leveraging Temporal Point Processes (TPPs) to identify abnormal temporal activity and using LLM-generated scores to refine detection. This method aims to overcome limitations of existing supervised and unsupervised techniques by not relying on oversimplified assumptions of user coordination. Experiments on five real-world IO datasets demonstrated TENSOR's superior performance compared to baseline methods. AI
IMPACT This research could lead to more effective tools for identifying and mitigating coordinated disinformation campaigns on social media.
RANK_REASON The cluster contains a research paper detailing a new method for anomaly detection in AI.
- alphaXiv
- arXiv
- CatalyzeX
- DagsHub
- Gotit.pub
- Hugging Face
- IArxiv
- LLM
- ScienceCast
- Temporal Point Process
- TENSOR
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
