{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/164851"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/164851","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"Efficient Uncertainty Quantification of Large Language Models","abstract":"Large Language Models (LLMs) have demonstrated remarkable success in many applications; however, their reliability remains a critical concern, especially in high-stakes domains such as healthcare, finance, and law. Uncertainty Quantification (UQ) is essential for assessing LLM outputs and ensuring trust. However, existing UQ methods for LLMs face challenges: high computational costs, difficulties in handling unstructured outputs, and limited generalizability. This thesis addresses these challenges by proposing a systematic investigation into robust and efficient UQ methodologies tailored for LLMs. Specifically, this work focuses on: (1) analyzing probing methods to determine how hidden layers encode information relevant to uncertainty and accuracy, (2) developing novel UQ metrics that strongly correlate with actual model performance, and (3) designing computationally efficient pipelines to make UQ practical for real-world applications. By bridging these gaps, this research aims to establish UQ as a reliable tool for evaluating and improving the trustworthiness of LLM outputs, facilitating their safe and effective deployment in critical domains.","abstract_html":"Large Language Models (LLMs) have demonstrated remarkable success in many applications; however, their reliability remains a critical concern, especially in high-stakes domains such as healthcare, finance, and law. Uncertainty Quantification (UQ) is essential for assessing LLM outputs and ensuring trust. However, existing UQ methods for LLMs face challenges: high computational costs, difficulties in handling unstructured outputs, and limited generalizability. This thesis addresses these challenges by proposing a systematic investigation into robust and efficient UQ methodologies tailored for LLMs. Specifically, this work focuses on: (1) analyzing probing methods to determine how hidden layers encode information relevant to uncertainty and accuracy, (2) developing novel UQ metrics that strongly correlate with actual model performance, and (3) designing computationally efficient pipelines to make UQ practical for real-world applications. By bridging these gaps, this research aims to establish UQ as a reliable tool for evaluating and improving the trustworthiness of LLM outputs, facilitating their safe and effective deployment in critical domains.","abstract_has_math":false,"creators":["Li, Angela"],"institution":"Massachusetts Institute of Technology","degree_name":"Master","degree_level":null,"degree_discipline":null,"degree_department":"Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","school":null,"contributors":[],"advisors":["Ghassemi, Marzyeh"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-09","date_published":"2025-09","updated_at":"2026-07-22T22:21:45Z","subjects":[],"languages":[],"rights":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"],"rights_urls":["https://rightsstatements.org/page/InC-EDU/1.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/164851","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Ghassemi, Marzyeh"]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"]},{"key":"dc:creator","label":"Author","values":["Li, Angela"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-02-12T17:14:32Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-02-12T17:14:32Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-09"]},{"key":"dc:publisher","label":"Institution","values":["Massachusetts Institute of Technology"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://rightsstatements.org/page/InC-EDU/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/164851"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Large Language Models (LLMs) have demonstrated remarkable success in many applications; however, their reliability remains a critical concern, especially in high-stakes domains such as healthcare, finance, and law. Uncertainty Quantification (UQ) is essential for assessing LLM outputs and ensuring trust. However, existing UQ methods for LLMs face challenges: high computational costs, difficulties in handling unstructured outputs, and limited generalizability. This thesis addresses these challenges by proposing a systematic investigation into robust and efficient UQ methodologies tailored for LLMs. Specifically, this work focuses on: (1) analyzing probing methods to determine how hidden layers encode information relevant to uncertainty and accuracy, (2) developing novel UQ metrics that strongly correlate with actual model performance, and (3) designing computationally efficient pipelines to make UQ practical for real-world applications. By bridging these gaps, this research aims to establish UQ as a reliable tool for evaluating and improving the trustworthiness of LLM outputs, facilitating their safe and effective deployment in critical domains."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["MNG"]},{"key":"dc:title","label":"Title","values":["Efficient Uncertainty Quantification of Large Language Models"]}]}],"canonical_facts":{"dc:contributor.advisor":["Ghassemi, Marzyeh"],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"],"dc:creator":["Li, Angela"],"dc:date.accessioned":["2026-02-12T17:14:32Z"],"dc:date.available":["2026-02-12T17:14:32Z"],"dc:date.issued":["2025-09"],"dc:description.abstract":["Large Language Models (LLMs) have demonstrated remarkable success in many applications; however, their reliability remains a critical concern, especially in high-stakes domains such as healthcare, finance, and law. Uncertainty Quantification (UQ) is essential for assessing LLM outputs and ensuring trust. However, existing UQ methods for LLMs face challenges: high computational costs, difficulties in handling unstructured outputs, and limited generalizability. This thesis addresses these challenges by proposing a systematic investigation into robust and efficient UQ methodologies tailored for LLMs. Specifically, this work focuses on: (1) analyzing probing methods to determine how hidden layers encode information relevant to uncertainty and accuracy, (2) developing novel UQ metrics that strongly correlate with actual model performance, and (3) designing computationally efficient pipelines to make UQ practical for real-world applications. By bridging these gaps, this research aims to establish UQ as a reliable tool for evaluating and improving the trustworthiness of LLM outputs, facilitating their safe and effective deployment in critical domains."],"dc:description.degree":["MNG"],"dc:identifier.uri":["https://hdl.handle.net/1721.1/164851"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"],"dc:rights.uri":["https://rightsstatements.org/page/InC-EDU/1.0/"],"dc:title":["Efficient Uncertainty Quantification of Large Language Models"],"dc:type":["Thesis"],"thesis:degree_name":["Master"]},"updated_at":"2026-07-22T22:21:45Z"}