# DariMis: Harm-Aware Modeling for Dari Misinformation Detection on YouTube

arXiv: https://arxiv.org/abs/2603.22977 · HTML: https://mlsanity.com/paper/2603.22977/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 45 references that are in the map.
- First posted on arXiv 24 Mar 2026.

## Builds on (most cited first, 24 of 24)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [FEVER: a large-scale dataset for Fact Extraction and VERification](https://mlsanity.com/paper/1803.05355/) (2018)
- [The State and Fate of Linguistic Diversity and Inclusion in the NLP World](https://mlsanity.com/paper/2004.09095/) (2020)
- [XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization](https://mlsanity.com/paper/2003.11080/) (2020)
- [A Survey on Automated Fact-Checking](https://mlsanity.com/paper/2108.11896/) (2021)
- ["Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection](https://mlsanity.com/paper/1705.00648/) (2017)
- [AraBERT: Transformer-based Model for Arabic Language Understanding](https://mlsanity.com/paper/2003.00104/) (2020)
- [Fake News Detection on Social Media: A Data Mining Perspective](https://mlsanity.com/paper/1708.01967/) (2017)
- [Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotations](https://mlsanity.com/paper/2110.05719/) (2021)
- [XGLUE: A New Benchmark Dataset for Cross-lingual Pre-training, Understanding and Generation](https://mlsanity.com/paper/2004.01401/) (2020)
- [Are All Languages Created Equal in Multilingual BERT?](https://mlsanity.com/paper/2005.09093/) (2020)
- [MultiFC: A Real-World Multi-Domain Dataset for Evidence-Based Fact Checking of Claims](https://mlsanity.com/paper/1909.03242/) (2019)
- [MasakhaNER: Named Entity Recognition for African Languages](https://mlsanity.com/paper/2103.11811/) (2021)
- [Combining Fact Extraction and Verification with Neural Semantic Matching Networks](https://mlsanity.com/paper/1811.07039/) (2018)
- [X-FACTR: Multilingual Factual Knowledge Retrieval from Pretrained Language Models](https://mlsanity.com/paper/2010.06189/) (2020)
- [Detection and Resolution of Rumours in Social Media: A Survey](https://mlsanity.com/paper/1704.00656/) (2017)
- [DeClarE: Debunking Fake News and False Claims using Evidence-Aware Deep Learning](https://mlsanity.com/paper/1809.06416/) (2018)
- [The science of fake news](https://mlsanity.com/paper/2307.07903/) (2023)
- [ParsBERT: Transformer-based Model for Persian Language Understanding](https://mlsanity.com/paper/2005.12515/) (2020)
- [We Need to Talk About Random Splits](https://mlsanity.com/paper/2005.00636/) (2020)
- [Survey of Computational Approaches to Lexical Semantic Change](https://mlsanity.com/paper/1811.06278/) (2018)
- [Auditing Radicalization Pathways on YouTube](https://mlsanity.com/paper/1908.08313/) (2019)
- [Who Funds Misinformation? A Systematic Analysis of the Ad-related Profit Routines of Fake News sites](https://mlsanity.com/paper/2202.05079/) (2022)

## Abstract (arXiv)

Dari, the primary language of Afghanistan, is spoken by tens of millions of people yet remains largely absent from the misinformation detection literature. We address this gap with DariMis, the first manually annotated dataset of 9,224 Dari-language YouTube videos, labeled across two dimensions: Information Type (Misinformation, Partly True, True) and Harm Level (Low, Medium, High). A central empirical finding is that these dimensions are structurally coupled, not independent: 55.9 percent of Misinformation carries at least Medium harm potential, compared with only 1.0 percent of True content. This enables Information Type classifiers to function as implicit harm-triage filters in content moderation pipelines. We further propose a pair-input encoding strategy that represents the video title and description as separate BERT segment inputs, explicitly modeling the semantic relationship between headline claims and body content, a key signal of misleading information. An ablation study against single-field concatenation shows that pair-input encoding yields a 7.0 percentage point gain in Misinformation recall (60.1 percent to 67.1 percent), the safety-critical minority class, despite modest overall macro F1 differences (0.09 percentage points). We benchmark a Dari/Farsi-specialized model (ParsBERT) against XLM-RoBERTa-base; ParsBERT achieves the best test performance with accuracy of 76.60 percent and macro F1 of 72.77 percent. Bootstrap 95 percent confidence intervals are reported for all metrics, and we discuss both the practical significance and statistical limitations of the results.