[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-273647-105":59,"doc-detail-273647-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","automated-summarization-of-financial-news-using-large-language-models-and-retrieval-augmented-generation-an-early-empirical-study","Automated Summarization of Financial News Using Large Language Models and Retrieval-Augmented Generation - An Early Empirical Study","","Stock market analysts and investors face a daily challenge: too much financial news, too little time, and manual synthesis across hundreds of company articles is impractical while omissions can directly affect decisions. The study from George Washington University (Fall 2023) evaluates whether large language models can reliably automate summarization by combining news retrieval, company background from Wikipedia, and stock prices from Yahoo Finance, including a structured-to-text template and both Summarize Chains and RAG approaches.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/automated-summarization-of-financial-news-using-large-language-models-and-retrieval-augmented-generation-an-early-empirical-study/273647/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/automated-summarization-of-financial-news-using-large-language-models-and-retrieval-augmented-generation-an-early-empirical-study/273647.png","ImageObject",300,407,{"name":92,"@type":93},"Riley","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-21","2026-09-15",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",14,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does the project address in financial news workflows?","Question",{"text":113,"@type":114},"Analysts and investors must process too many news articles in too little time, and manual reading and synthesis can cause missed key information that affects investment decisions.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How is the summarization pipeline constructed in the study?",{"text":118,"@type":114},"It pulls news from News API, company background from Wikipedia, and stock prices from Yahoo Finance for ten major companies. It also converts numeric stock data into natural-language narratives using a structured-to-text template.",{"name":120,"@type":111,"acceptedAnswer":121},"Which summarization approach and model performed best, and why?",{"text":122,"@type":114},"Falcon-7B with Summarize Chains delivered the best results, producing accurate and coherent coverage. RAG showed issues such as repetition and hallucinated facts in some configurations.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},273647,1789472364,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":130,"read_time":145,"preview_page_count":4},1374391975076,"https://ap-avatar.wpscdn.com/avatar/14000253ca4ec9f6853?x-image-process=image/resize,m_fixed,w_180,h_180&k=1783305029341752051","arXiv : 2608 . 19526v 1 [cs .CL] 20 Aug 2026  \nAutomated Summarization of Financial News Using Large Language Models and Retrieval-Augmented Generation:  \nAn Early Empirical Study (Fall 2023) ∗  \nPranav Chandaliya  \nDepartment of Data Science  \nGeorge Washington University  \nWashington, D.C., USA  \n[pranavc@gwu. edu](pranavc@gwu. edu)  \nOctober 2023  \n(Manuscript prepared for public release, 2026)  \nAbstract  \nStock market analysts and investors face a daily challenge: too much financial news, too little time. Manually reading and synthesizing hundreds of company-specific articles is impractical, yet missing key information can directly affect investment decisions. This project, conducted at George Washington University in Fall 2023, explores whether Large Language Models can automate this process reliably.  \nWe built a pipeline that pulls news articles from the News API, company background from Wikipedia, and stock price data from Yahoo Finance for ten major companies (AAPL, MSFT, GOOGL, AMZN, META, TSLA, JPM, NVDA, WMT, DIS) . Because LLMs cannot directly process numerical tables, we developed a simple but effective template that converts stock data into natural language narratives. We then tested two summarization approaches (Summarize Chains and Retrieval-Augmented Generation (RAG) with FAISS) across three open-source models (Falcon-7B-Instruct, DistilBART-CNN-12-6, BART-Large-XSum) for news, and GPT (text-davinci-003) for stock summaries.  \nFalcon-7B with Summarize Chains gave the best results, covering all news events accurately and coherently. RAG, while promising in theory, caused severe repetition in Falcon and hallucinated facts in BART-Large when k was large. Both LLM-based approaches outperformed a simple Lead-3 baseline on ROUGE-1 . We also built a Streamlit dashboard for interactive stock visualization. The work was done in Fall 2023, before RAG-based financial tools became widespread, and the failure modes we document, particularly hallucination under RAG in smaller models, remain relevant today.  \nKeywords: Large Language Models, Financial News Summarization, Retrieval-Augmented Generation, FAISS, Natural Language Processing, Stock Market Analysis, Falcon-7B, Streamlit  \nContents  \n1 Introduction 3  \n∗ This research was conducted as part of DATS 6499: Applied Data Science Research at George Washington University, Fall 2023 . The author thanks Prof. Vineet Bhagwat (School of Business, George Washington University) for his supervision and guidance throughout the project.  \n2 Related Work 3  \n2. 1 Text Summarization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3  \n2.2 Financial NLP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3  \n2.3 Data-to-Text Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4  \n2.4 Retrieval-Augmented Generation ........................... 4  \n3 Data Collection and Preprocessing 4  \n3. 1 News Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4  \n3.2 Wikipedia Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4  \n3.3 Stock Market Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5  \n3.4 Feeding Numerical Data to LLMs: The Structured-to-Text Approach ....... 5  \n3.5 Knowledge Base Construction ............................. 6  \n3.6 Dataset Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6  \n4 Methodology 6  \n4.1 System Architecture .................................. 6  \n4.2 Text Chunking and Embedding ............................ 7  \n4.3 Summarize Chains Approach ............................. 7  \n4.4 Retrieval-Augmented Generation (RAG) Approach ................. 7  \n4.5 Language Models Evaluated . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8  \n4.6 GPT-Based Stock Summarization . . . . . . . . . . . . . . . . . . . . . . . . . . . 8  \n4.7 Interactive Dashboard . . . . . . . . . . . . .","cbCaitiFmfldyUEd","https://ap.wps.com/l/cbCaitiFmfldyUEd","pdf",324862,15,"English","# Introduction\n# Related Work\n# Data Collection and Preprocessing\n## Feeding Numerical Data to LLMs: The Structured-to-Text Approach\n## Knowledge Base Construction\n# Methodology\n## System Architecture\n## Summarize Chains Approach\n## Retrieval-Augmented Generation (RAG) Approach\n# Experiments and Results\n## Quantitative Evaluation: ROUGE Scores\n## Generated Financial Insights\n# Discussion\n# Conclusion","[{\"question\":\"What problem does the project address in financial news workflows?\",\"answer\":\"Analysts and investors must process too many news articles in too little time, and manual reading and synthesis can cause missed key information that affects investment decisions.\"},{\"question\":\"How is the summarization pipeline constructed in the study?\",\"answer\":\"It pulls news from News API, company background from Wikipedia, and stock prices from Yahoo Finance for ten major companies. It also converts numeric stock data into natural-language narratives using a structured-to-text template.\"},{\"question\":\"Which summarization approach and model performed best, and why?\",\"answer\":\"Falcon-7B with Summarize Chains delivered the best results, producing accurate and coherent coverage. RAG showed issues such as repetition and hallucinated facts in some configurations.\"}]","Automated Summarization of Financial News Using Large Language Models and Retrieval-Augmented Generation - An Early Empirical Study | PDF",38]