Language models are domain-specific chart analysts
As the advancement of multi-modal Large Language Models (LLM) such as GPT4, the cognitive capability of models is facing new expectations. Meanwhile, when LLM trainings are getting more expensive, there has been a gap between the conventional pretrain-finetune paradigm and the LLM prompting paradigm...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/167416 |
| _version_ | 1826125796667817984 |
|---|---|
| author | Zhao, Yinjie |
| author2 | Wen Bihan |
| author_facet | Wen Bihan Zhao, Yinjie |
| author_sort | Zhao, Yinjie |
| collection | NTU |
| description | As the advancement of multi-modal Large Language Models (LLM) such as GPT4, the cognitive capability of models is facing new expectations. Meanwhile, when LLM trainings are getting more expensive, there has been a gap between the conventional pretrain-finetune paradigm and the LLM prompting paradigm regarding model designing. In order to close the currently existing gaps, we propose an AI model engineering pipeline, Cost-efficient C2T Pipeline (C2P), towards an objective of C2T model cognitive capabilities on Chart Domain-specific Analyzing (CDA). A 41.5 million parameter model was trained under C2P, achieving a significantly higher cost-efficiency compared to other models, with a comparable performance. In order to conduct the experiment validation, we proposed a new dataset, EconCharts, which is a domain-specific dataset on economics. C2P explores the Domain-specific cognitive capabilities of C2T / LLM models and to fill the engineering gap between expensive LLM models together with lightweight C2T models. |
| first_indexed | 2024-10-01T06:42:33Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/167416 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:42:33Z |
| publishDate | 2023 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1674162023-07-07T15:43:37Z Language models are domain-specific chart analysts Zhao, Yinjie Wen Bihan School of Electrical and Electronic Engineering bihan.wen@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence As the advancement of multi-modal Large Language Models (LLM) such as GPT4, the cognitive capability of models is facing new expectations. Meanwhile, when LLM trainings are getting more expensive, there has been a gap between the conventional pretrain-finetune paradigm and the LLM prompting paradigm regarding model designing. In order to close the currently existing gaps, we propose an AI model engineering pipeline, Cost-efficient C2T Pipeline (C2P), towards an objective of C2T model cognitive capabilities on Chart Domain-specific Analyzing (CDA). A 41.5 million parameter model was trained under C2P, achieving a significantly higher cost-efficiency compared to other models, with a comparable performance. In order to conduct the experiment validation, we proposed a new dataset, EconCharts, which is a domain-specific dataset on economics. C2P explores the Domain-specific cognitive capabilities of C2T / LLM models and to fill the engineering gap between expensive LLM models together with lightweight C2T models. Bachelor of Engineering (Electrical and Electronic Engineering) 2023-05-28T11:04:56Z 2023-05-28T11:04:56Z 2023 Final Year Project (FYP) Zhao, Y. (2023). Language models are domain-specific chart analysts. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/167416 https://hdl.handle.net/10356/167416 en A3271-221 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Zhao, Yinjie Language models are domain-specific chart analysts |
| title | Language models are domain-specific chart analysts |
| title_full | Language models are domain-specific chart analysts |
| title_fullStr | Language models are domain-specific chart analysts |
| title_full_unstemmed | Language models are domain-specific chart analysts |
| title_short | Language models are domain-specific chart analysts |
| title_sort | language models are domain specific chart analysts |
| topic | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence |
| url | https://hdl.handle.net/10356/167416 |
| work_keys_str_mv | AT zhaoyinjie languagemodelsaredomainspecificchartanalysts |