Instruction-Tuning For Chart Comprehension And Reasoning
dc.contributor.advisor | Prince, Emmanuel Hoque | |
dc.contributor.author | Shah Mohammadi, Mehrad | |
dc.date.accessioned | 2025-04-10T10:49:19Z | |
dc.date.available | 2025-04-10T10:49:19Z | |
dc.date.copyright | 2024-11-15 | |
dc.date.issued | 2025-04-10 | |
dc.date.updated | 2025-04-10T10:49:18Z | |
dc.degree.discipline | Information Systems and Technology | |
dc.degree.level | Master's | |
dc.degree.name | MA - Master of Arts | |
dc.description.abstract | Charts provide visual representations of data and are widely used for analyzing information, addressing queries, and conveying insights to others. Various chart-related downstream tasks have emerged recently, such as questionanswering and summarization. A common strategy to solve these tasks is to fine-tune various models originally trained on vision tasks language. However, such task-specific models are not capable of solving a wide range of chartrelated tasks, constraining their real-world applicability. To overcome these challenges, we introduce ChartInstruct: a novel chart-specific vision-language Instruction-following dataset comprising if instructions generated with distinct charts. We then present two distinct systems for instruction tuning on such datasets: (1) an end-to-end model (2) a pipeline model employing a two-step approach. Evaluation shows that our instruction-tuning approach supports a wide array of real-world chart comprehension and reasoning scenarios, thereby expanding the scope and applicability of our models to new kinds of tasks. | |
dc.identifier.uri | https://hdl.handle.net/10315/42807 | |
dc.language | en | |
dc.rights | Author owns copyright, except where explicitly noted. Please contact the author directly with licensing requests. | |
dc.subject.keywords | Natural language processing | |
dc.subject.keywords | Multimodal learning | |
dc.subject.keywords | Charts | |
dc.subject.keywords | Computer vision | |
dc.subject.keywords | Deep learning | |
dc.title | Instruction-Tuning For Chart Comprehension And Reasoning | |
dc.type | Electronic Thesis or Dissertation |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- Shah_Mohammadi_Mehrad_2024_MA.pdf
- Size:
- 5.22 MB
- Format:
- Adobe Portable Document Format