Papers by Kim Tran
A Joint Learning Approach based on Self-Distillation for Keyphrase Extraction from Scientific Documents (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods for keyphrase extraction are limited by the number of annotated documents. |
| Approach: | They propose a joint learning approach that uses the idea of self-distillation to extract keyphrases from unlabeled articles. |
| Outcome: | The proposed approach outperforms baseline models on two public benchmarks: Inspec and SemEval-2017. |
Z-GMOT: Zero-shot Generic Multiple Object Tracking (2024.findings-naacl)
Copied to clipboard
Kim Tran, Anh Duy Le Dinh, Tien-Phat Nguyen, Thinh Phan, Pha Nguyen, Khoa Luu, Donald Adjeroh, Gianfranco Doretto, Ngan Le
| Challenge: | Existing approaches to Multi-Object Tracking (MOT) rely on initial bounding boxes and struggle with unseen objects. |
| Approach: | They propose a cutting-edge multi-object tracking solution that can track unseen objects . they propose iGLIP and MA-SORT, which integrate motion and appearance matching strategies . |
| Outcome: | The proposed solution can track objects from never-seen categories without initial bounding boxes or predefined categories. |
Fine-tuning CLIP Text Encoders with Two-step Paraphrasing (2024.findings-eacl)
Copied to clipboard
| Challenge: | Contrastive language-image pre-training models have demonstrated considerable success across various vision-language tasks, such as text-to-image retrieval. |
| Approach: | They propose a fine-tuning approach to enhance the representations of CLIP models for paraphrases by leveraging large language models. |
| Outcome: | The proposed model improves on baseline models across paraphrased retrieval, visual genome relation and attribution, and seven semantic textual similarity tasks. |
When and Why is Document-level Context Useful in Neural Machine Translation? (D19-65)
Copied to clipboard
| Challenge: | Recent advances in document-level NMT focus on sophisticated integration of the context, explaining its improvement with only a few selected examples or targeted test sets. |
| Approach: | They extensively quantify the causes of improvements by a document-level model in general test sets, clarifying the limit of the usefulness of document- level context in NMT. |
| Outcome: | The proposed model is not interpretable as utilizing the context, and a long context is not helpful for NMT. |