Papers by Mladen Karan
Classification-Based Self-Learning for Weakly Supervised Bilingual Lexicon Induction (2020.acl-main)
Copied to clipboard
| Challenge: | Effective projection-based cross-lingual word embedding induction relies on the iterative self-learning procedure. |
| Approach: | They propose a classification-based approach to self-learning that allows for integration of diverse features into the iterative process. |
| Outcome: | The proposed method improves bilingual lexicon induction on a weakly supervised setup with 28 language pairs. |
XHate-999: Analyzing and Detecting Abusive Language Across Domains and Languages (2020.coling-main)
Copied to clipboard
| Challenge: | XHate-999 is a multi-domain and multilingual evaluation data set for abusive language detection . we show that domain- and language-adaption can lead to substantially improved abusive language detecting in the target language . |
| Approach: | They propose a multi-domain and multilingual evaluation data set for abusive language detection that allows for disentanglement of domain transfer and language transfer effects. |
| Outcome: | The proposed model can significantly improve abusive language detection in the target language in the zero-shot transfer setups. |
CoRAL: a Context-aware Croatian Abusive Language Dataset (2022.findings-aacl)
Copied to clipboard
| Challenge: | Semi-automated comment moderation systems can greatly aid human moderators by either automatically classifying the examples or allowing the moderator to prioritize which comments to consider first. |
| Approach: | They propose to use a language and culturally aware Croatian Abusive dataset to analyze inappropriate comments in a context-based manner. |
| Outcome: | The proposed dataset shows that current models degrade when comments are not explicit and further degrades when language skill and context knowledge are required to interpret the comment. |
LEDA: a Large-Organization Email-Based Decision-Dialogue-Act Analysis Dataset (2023.findings-acl)
Copied to clipboard
Mladen Karan, Prashant Khare, Ravi Shekhar, Stephen McQuistin, Ignacio Castro, Gareth Tyson, Colin Perkins, Patrick Healey, Matthew Purver
| Challenge: | Using dialog acts to study decision-making in large distributed organizations is challenging due to the size and distributed nature of such groups. |
| Approach: | They propose a set of dialog acts for the study of decision-making mechanisms in large distributed organizations. |
| Outcome: | The proposed dataset can be used to better understand decision-making in large distributed organizations. |
Tracing Linguistic Markers of Influence in a Large Online Organisation (2023.acl-short)
Copied to clipboard
Prashant Khare, Ravi Shekhar, Mladen Karan, Stephen McQuistin, Colin Perkins, Ignacio Castro, Gareth Tyson, Patrick Healey, Matthew Purver
| Challenge: | Social science and psycholinguistic research have shown that power and status affect how people use language in a range of domains. |
| Approach: | They propose to use lexical categories and BERT to predict levels of influence in an online community and identify key linguistic differences between people before and after becoming influential. |
| Outcome: | The results show that participants' levels of influence can be predicted from their email text, and identify key differences in language use for the same person before and after becoming influential. |
Denoising Labeled Data for Comment Moderation Using Active Learning (2024.lrec-main)
Copied to clipboard
| Challenge: | Large contextualized language models (LLMs) are becoming ubiquitous in natural language processing due to their performance and adaptability to diverse tasks. |
| Approach: | They propose to use active learning methods to denoise textual data for model training by sampling the most informative examples with noisy labels with active learning. |
| Outcome: | The proposed method reduces the cost of reannotation by reducing noise in noisy examples. |