Published in EACL VarDial Workshop, 2023
A two-stage end-to-end deep learning based pipeline for multilingual dialect detection.
Recommended citation: Vaidya, A. and Kane, A., 2023. Two-stage Pipeline for Multilingual Dialect Detection. arXiv preprint arXiv:2303.03487. https://arxiv.org/pdf/2303.03487.pdf
Published in ICASSP, 2023
A state-of-the-art baseline for few shot out-of-distribution detection in NLP.
Recommended citation: C. Gautam, A. Kane, S. Ramasamy and S. Sundaram, "Unsupervised Out-of-Distribution Detection Using Few in-Distribution Samples," ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Rhodes Island, Greece, 2023, pp. 1-5, doi: 10.1109/ICASSP49357.2023.10096482. https://ieeexplore.ieee.org/abstract/document/10096482
Published in FIRE 2022 workshop, 2022
Applying LLMs to offensive language detection in Marathi.
Recommended citation: Chavan, T., Patankar, S., Kane, A., Gokhale, O. and Joshi, R., 2022. A Twitter BERT Approach for Offensive Language Detection in Marathi. arXiv preprint arXiv:2212.10039. https://arxiv.org/abs/2212.10039
Published in AAAI 2023 Deployable AI workshop, 2022
A state-of-the-art baseline for VQA on the FloodNet dataset. Won the Second Best Paper award in NewInML Workshop 2022.
Recommended citation: Kirtane, N., Manushree, V. and Kane, A., 2022. Efficient Gender Debiasing of Pre-trained Indic Language Models. arXiv preprint arXiv:2209.03661. https://arxiv.org/abs/2209.03661
Published in NeurIPS 2022 Tackling Climate Change with Machine Learning workshop', 2022
We study the VQA problem for the Floodnet dataset in continual and zero shot setting.
Recommended citation: Kane, A., Manushree, V. and Khose, S., 2022. Continual VQA for Disaster Response Systems. arXiv preprint arXiv:2209.10320. https://arxiv.org/abs/2209.10320
Published in EMNLP 2022 EvoNLP workshop, 2022
We study two methods for word sense disamiguation using temporal components.
Recommended citation: Godbole, M., Dandavate, P. and Kane, A., 2022. Temporal Word Meaning Disambiguation using TimeLMs. arXiv preprint arXiv:2210.08207. https://arxiv.org/abs/2210.08207
Published in EMNLP Arabic Natural Language Processing Workshop, 2022
Ablation of Arabic models for propoganda detection in Arabic.
Recommended citation: Chavan, Tanmay and kane, Aditya. “Large Language Models for Multi-label Propaganda Detection” (2022). https://arxiv.org/abs/2210.08209
Published in ICML NewInML Workshop, 2022
A state-of-the-art baseline for VQA on the FloodNet dataset. Won the Second Best Paper award in NewInML Workshop 2022.
Recommended citation: Kane, Aditya and Sahil Khose. “An Efficient Modern Baseline for FloodNet VQA.” (2022). https://arxiv.org/pdf/2205.15025.pdf
Published in ACL WASSA Workshop, 2022
Emotion detection using innovative data sampling techniques and ensemble of transformers.
Recommended citation: Aditya Kane, Shantanu Patankar, Sahil Khose, and Neeraja Kirtane. 2022. Transformer based ensemble for emotion detection. In Proceedings of the 12th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis, pages 250–254, Dublin, Ireland. Association for Computational Linguistics. https://aclanthology.org/2022.wassa-1.25/