Papers by Ifeoluwa Wuraola
SLANG-GraphRAG: Multi-Layered Retrieval with Domain-Specific Knowledge for Low Resource Social Media Conversations (2026.findings-eacl)
Copied to clipboard
| Challenge: | Standard NLP benchmarks often miss subtle, culturally-specific cues in social media . incorporating structured cultural knowledge into the retrieval process improves accuracy by up to 31% . |
| Approach: | They propose a retrieval-augmented framework that integrates a culture-specific slang knowledge graph into large language models via one-shot prompting. |
| Outcome: | The proposed framework outperforms traditional and unstructured retrieval methods in slang-based models by 31% and 28%. |
Understanding Slang with LLMs: Modelling Cross-Cultural Nuances through Paraphrasing (2024.emnlp-main)
Copied to clipboard
| Challenge: | a recent study examines the ability of large language models (LLMs) to paraphrase slang within climate-related tweets . slanted tweets from non-anglocentric countries may contain cultural references and idioms based on sociocultural identities . |
| Approach: | They investigate the ability of large language models to paraphrase slang within climate-related tweets from Nigeria and the UK. |
| Outcome: | The proposed model can paraphrase slang within climate-related tweets from Nigeria and the UK . the model can only parse sexist and sex-related slurs, the study shows . |