# Hidden Persuasion: Detecting Manipulative Narratives on Social Media During the 2022 Russian Invasion of Ukraine

arXiv: https://arxiv.org/abs/2505.24028 · HTML: https://mlsanity.com/paper/2505.24028/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 16 references that are in the map.
- First posted on arXiv 29 May 2025.

## Builds on (most cited first, 9 of 9)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders](https://mlsanity.com/paper/2404.05961/) (2024)
- [mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval](https://mlsanity.com/paper/2407.19669/) (2024)
- [SemEval-2020 Task 11: Detection of Propaganda Techniques in News Articles](https://mlsanity.com/paper/2009.02696/) (2020)
- [WikiCheck: An end-to-end open source Automatic Fact-Checking API based on Wikipedia](https://mlsanity.com/paper/2109.00835/) (2021)
- [LLM-RM at SemEval-2023 Task 2: Multilingual Complex NER using XLM-RoBERTa](https://mlsanity.com/paper/2305.03300/) (2023)
- [An Open Multilingual System for Scoring Readability of Wikipedia](https://mlsanity.com/paper/2406.01835/) (2024)
- [From Bytes to Borsch: Fine-Tuning Gemma and Mistral for the Ukrainian Language Representation](https://mlsanity.com/paper/2404.09138/) (2024)
- [Characterizing Knowledge Manipulation in a Russian Wikipedia Fork](https://mlsanity.com/paper/2504.10663/) (2025)

## Abstract (arXiv)

This paper presents one of the top-performing solutions to the UNLP 2025 Shared Task on Detecting Manipulation in Social Media. The task focuses on detecting and classifying rhetorical and stylistic manipulation techniques used to influence Ukrainian Telegram users. For the classification subtask, we fine-tuned the Gemma 2 language model with LoRA adapters and applied a second-level classifier leveraging meta-features and threshold optimization. For span detection, we employed an XLM-RoBERTa model trained for multi-target, including token binary classification. Our approach achieved 2nd place in classification and 3rd place in span detection.