Papers by Gizem Gezici
Seeing All Sides: Multi-Perspective In-Context Learning for Subjective NLP (2026.findings-eacl)
Copied to clipboard
| Challenge: | Modern language models excel at factual reasoning but struggle with value diversity, authors say . task-sensitive tasks such as hate speech expose this limitation . human disagreement captures the diversity of plausible human perspectives, authors argue . |
| Approach: | They evaluate four large language models with human disagreements on five datasets . they find multi-perspective in-context learning outperforms standard prompting . |
| Outcome: | The proposed approach outperforms standard prompting on English labels while disaggregated soft predictions better align with human judgments in Arabic and Italian datasets. |
#Turki$hTweets: A Benchmark Dataset for Turkish Text Correction (2020.findings-emnlp)
Copied to clipboard
| Challenge: | #Turki$hTweets provides correct/incorrect word annotations with a detailed misspelling category formulation based on real user data. |
| Approach: | #Turki$hTweets is a benchmark dataset for the task of correcting the user misspellings. |
| Outcome: | The proposed dataset is the first public dataset in this area. |