Enhancing agronomic advisory in East Africa through integration of domain- Adapted Large Language Models (LLMs) and existing data-driven frameworks.
This study explores how integration of domain-adapted Large Language Models (LLMs) and existing data-driven frameworks enhance wheat agronomic advisory services in east Africa. Although LLMs hold transformative potential across various domains, their use in agro-advisory services in resource-constra...
| Main Author: | |
|---|---|
| Format: | Tesis |
| Language: | Inglés |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10568/176138 |
| _version_ | 1855540560554622976 |
|---|---|
| author | Tilaye, A. |
| author_browse | Tilaye, A. |
| author_facet | Tilaye, A. |
| author_sort | Tilaye, A. |
| collection | Repository of Agricultural Research Outputs (CGSpace) |
| description | This study explores how integration of domain-adapted Large Language Models (LLMs) and existing data-driven frameworks enhance wheat agronomic advisory services in east Africa. Although LLMs hold transformative potential across various domains, their use in agro-advisory services in resource-constrained regions like East Africa is limited by a lack of domain-specific data and inadequate infrastructure. The general objective of this study is develop domain-adapted large language models (LLMs) that integrate heterogeneous agronomic inputs with existing data-driven frameworks, and evaluate their performance, and cross-regional adaptability for delivering accurate and context-aware agro-advisories. By extracting and structuring heterogeneous inputs from diverse sources using LLMs an agronomic knowledge base will be developed. Then open-source LLMs via efficient techniques (e.g., QLoRA), will be fine-tuned and integrated with existing geospatially enabled advisory systems using Retrieval-Augmented Generation (RAG). Performance of LLMs will be assessed through quantitative metrics and expert validation. Cross-regional adaptability will be also tested by using in-context learning and evaluated using expert validation. Expected outcomes include efficient pipeline that extracts and curates agronomic information from diverse sources, alongside a domain-adapted LLM integrated with existing data‐driven frameworks. Finally findings of the study will be disseminated via publications and presentations |
| format | Tesis |
| id | CGSpace176138 |
| institution | CGIAR Consortium |
| language | Inglés |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| record_format | dspace |
| spelling | CGSpace1761382025-08-26T01:11:36Z Enhancing agronomic advisory in East Africa through integration of domain- Adapted Large Language Models (LLMs) and existing data-driven frameworks. Tilaye, A. large language models east africa frameworks integration agronomy This study explores how integration of domain-adapted Large Language Models (LLMs) and existing data-driven frameworks enhance wheat agronomic advisory services in east Africa. Although LLMs hold transformative potential across various domains, their use in agro-advisory services in resource-constrained regions like East Africa is limited by a lack of domain-specific data and inadequate infrastructure. The general objective of this study is develop domain-adapted large language models (LLMs) that integrate heterogeneous agronomic inputs with existing data-driven frameworks, and evaluate their performance, and cross-regional adaptability for delivering accurate and context-aware agro-advisories. By extracting and structuring heterogeneous inputs from diverse sources using LLMs an agronomic knowledge base will be developed. Then open-source LLMs via efficient techniques (e.g., QLoRA), will be fine-tuned and integrated with existing geospatially enabled advisory systems using Retrieval-Augmented Generation (RAG). Performance of LLMs will be assessed through quantitative metrics and expert validation. Cross-regional adaptability will be also tested by using in-context learning and evaluated using expert validation. Expected outcomes include efficient pipeline that extracts and curates agronomic information from diverse sources, alongside a domain-adapted LLM integrated with existing data‐driven frameworks. Finally findings of the study will be disseminated via publications and presentations 2025-06-25 2025-08-19T07:17:11Z 2025-08-19T07:17:11Z Thesis https://hdl.handle.net/10568/176138 en Open Access application/pdf Tilaye, A. ( 2025). Enhancing agronomic advisory in East Africa through integration of domain- Adapted Large Language Models (LLMs) and existing data-driven frameworks. University Mohammed VI polytechnic. 52p. |
| spellingShingle | large language models east africa frameworks integration agronomy Tilaye, A. Enhancing agronomic advisory in East Africa through integration of domain- Adapted Large Language Models (LLMs) and existing data-driven frameworks. |
| title | Enhancing agronomic advisory in East Africa through integration of domain- Adapted Large Language Models (LLMs) and existing data-driven frameworks. |
| title_full | Enhancing agronomic advisory in East Africa through integration of domain- Adapted Large Language Models (LLMs) and existing data-driven frameworks. |
| title_fullStr | Enhancing agronomic advisory in East Africa through integration of domain- Adapted Large Language Models (LLMs) and existing data-driven frameworks. |
| title_full_unstemmed | Enhancing agronomic advisory in East Africa through integration of domain- Adapted Large Language Models (LLMs) and existing data-driven frameworks. |
| title_short | Enhancing agronomic advisory in East Africa through integration of domain- Adapted Large Language Models (LLMs) and existing data-driven frameworks. |
| title_sort | enhancing agronomic advisory in east africa through integration of domain adapted large language models llms and existing data driven frameworks |
| topic | large language models east africa frameworks integration agronomy |
| url | https://hdl.handle.net/10568/176138 |
| work_keys_str_mv | AT tilayea enhancingagronomicadvisoryineastafricathroughintegrationofdomainadaptedlargelanguagemodelsllmsandexistingdatadrivenframeworks |