Beyond fact-checking: harnessing retrieval-augmented generation and large language models for tackling financial misinformation
Carucci, Matteo (A.A. 2023/2024) Beyond fact-checking: harnessing retrieval-augmented generation and large language models for tackling financial misinformation. Tesi di Laurea in Artificial intelligence and machine learning, Luiss Guido Carli, relatore Giuseppe Francesco Italiano, pp. 77. [Bachelor's Degree Thesis]
|
PDF (Full text)
Download (1MB) | Preview |
Abstract/Index
The pillars of financial stability. The economic impact of misinformation. The importance of accurate financial information and the role of journalism. The evolution of financial journalism. Literature review. Financial misinformation: definitions, types, and examples. The economic loss caused by financial misinformation. Current fact-checking methods: overview of manual and automated approaches. Large language models (LLMs): evolution, capabilities, and limitations. Retrieval augmented generation (RAG): introduction, how it works, and its advantages. The RAG architecture: a general overview. Retrieval augmented generation main advantages and shortcomings. RAG variations and a gentle introduction to agentic RAG. Methodology. A mixed design: combining self-RAG, adaptive and corrective RAG. Data collection: extracting financial news data through FN-SPID. Analytical framework: using RAG and LLMs (LangChain, and Llama3 etc.) for fact-checking. Comparative analysis: performance of RAG-enhanced LLMs versus traditional LLMs in identifying and correcting misinformation. Case study: application in stock market news. Insights and implications of our findings. Technical challenges: addressing RAG and LLM limitations.
References
Bibliografia: pp. 70-75.
| Thesis Type: | Bachelor's Degree Thesis |
|---|---|
| Institution: | Luiss Guido Carli |
| Degree Program: | Bachelor's Degree Programs > Bachelor's Degree Program in Management and Computer Science, English language (L-18) |
| Chair: | Artificial intelligence and machine learning |
| Thesis Supervisor: | Italiano, Giuseppe Francesco |
| Academic Year: | 2023/2024 |
| Session: | Summer |
| Deposited by: | Alessandro Perfetti |
| Date Deposited: | 16 Oct 2024 12:45 |
| Last Modified: | 16 Oct 2024 12:45 |
| URI: | https://tesi.luiss.it/id/eprint/40035 |
Downloads
Downloads per month over past year
Repository Staff Only
 |
View Item |
