2026-09-19T07:28:15Zhttps://dspace.mit.edu/server/oai/request

oai:dspace.mit.edu:1721.1/1646672026-01-30T03:24:54Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023

Cafarella, Michael J. Chen, Peter Baile Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science 2026-01-29T15:06:51Z 2026-01-29T15:06:51Z 2025-09 2025-09-15T14:39:56.131Z https://hdl.handle.net/1721.1/164667 Strong retrieval and reasoning capabilities are essential for large language models (LLMs) to effectively handle a broad spectrum of downstream tasks, such as open-domain question answering and solving math or science problems. While current LLM-based frameworks achieve strong performance on complex retrieval and reasoning tasks, they do so at a high computational cost. Additionally, they often lack structured, systematic problem-solving strategies, leading to unexpected failures. In particular, these models typically operate in an iterative, online, and isolated fashion—failing to exploit relationships across data sources, opportunities for offline computation, and the benefits of reusability—resulting in less-than-optimal outcomes. In contrast, traditional data management systems are engineered for both efficiency and accuracy, with careful coordination across all stages of the query pipeline. Inspired by these principles, this work proposes novel approaches to improve LLMbased retrieval and reasoning by incorporating optimization techniques from data systems. Our evaluation across a range of knowledge- and reasoning-intensive datasets demonstrates significant gains in both accuracy and computational efficiency. S.M. Massachusetts Institute of Technology In Copyright - Educational Use Permitted Copyright retained by author(s) https://rightsstatements.org/page/InC-EDU/1.0/ Optimizing Large Language Models from a Data SystemsPerspective Thesis application/pdf Master Master of Science in Electrical Engineering and Computer Science Publication unknown unknown <Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="9b306fa2-254c-4b00-a87c-296e028978ff"> <Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type> <Title>Optimizing Large Language Models from a Data SystemsPerspective</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2025-09</PublicationDate> <Authors> <Author> <DisplayName>Chen, Peter Baile</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <DisplayName>Massachusetts Institute of Technology</DisplayName> <OrgUnit /> </Publisher> </Publishers> <License>https://rightsstatements.org/page/InC-EDU/1.0/</License> <Abstract>Strong retrieval and reasoning capabilities are essential for large language models (LLMs) to effectively handle a broad spectrum of downstream tasks, such as open-domain question answering and solving math or science problems. While current LLM-based frameworks achieve strong performance on complex retrieval and reasoning tasks, they do so at a high computational cost. Additionally, they often lack structured, systematic problem-solving strategies, leading to unexpected failures. In particular, these models typically operate in an iterative, online, and isolated fashion—failing to exploit relationships across data sources, opportunities for offline computation, and the benefits of reusability—resulting in less-than-optimal outcomes. In contrast, traditional data management systems are engineered for both efficiency and accuracy, with careful coordination across all stages of the query pipeline. Inspired by these principles, this work proposes novel approaches to improve LLMbased retrieval and reasoning by incorporating optimization techniques from data systems. Our evaluation across a range of knowledge- and reasoning-intensive datasets demonstrates significant gains in both accuracy and computational efficiency.</Abstract> <Access xmlns="http://purl.org/coar/access_right" > </Access> </Publication>