서브메뉴
검색
Scaling Expertise Via Language Models: with Applications to Education
Scaling Expertise Via Language Models: with Applications to Education
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202103638
- ISBN
- 9798290625713
- DDC
- 306
- 서명/저자
- Scaling Expertise Via Language Models: with Applications to Education
- 발행사항
- [Sl] : Stanford University, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 283 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-01, Section: A.
- 주기사항
- Advisor: Demszky, Dora;Yang, Diyi.
- 학위논문주기
- Thesis (Ph.D.)--Stanford University, 2025.
- 초록/해제
- 요약Access to expertise shapes how individuals learn, develop, and succeed across society. For example, in education, experienced teachers teach students and train novice educators through effective interactions. However, access to expertise is limited, undermining learning at scale. While language models promise to democratize access, they often mimic surface-level patterns and lack the human touch needed to support learners through challenges. In this thesis, I present a new paradigm, scalable expertise, that reimagines how we can extend human-like expertise by adapting and leveraging language models for high-stakes, real-world interactions. In Part I, I introduce a computational method to extract expert reasoning from verbalized talk aloud protocols, enabling language models to handle complex tasks. In Part II, I introduce benchmarks aligned to domain needs and, resultingly, provide early insights into the reliability of language models for intervening on real-world interactions. In Part III, I introduce real-world applications of scaling expertise, including interventions with real-time AI assistance and methods to surface interaction patterns between real teachers and students. Finally, I conclude by discussing open questions and future directions for the paradigm of scalable expertise.The work presented in this thesis has broad implications: it demonstrates how language models---when adapted, evaluated and deployed appropriately---can enhance the human touch to improve real-world behaviors and outcomes at scale. While this thesis focuses on applications to education, it establishes the foundation for scaling expertise to other high-stakes domains like healthcare and law. By addressing the challenges of methods, evaluations, and applications, this thesis advances a human-centered vision of AI enhancing human learning at scale.
- 일반주제명
- Families & family life
- 일반주제명
- Tutoring
- 일반주제명
- Education
- 기타저자
- Stanford University.
- 기본자료저록
- Dissertations Abstracts International. 87-01A.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017358062
■00520260202103638
■006m o d
■007cr#unu||||||||
■020 ▼a9798290625713
■035 ▼a(MiAaPQ)AAI32090139
■035 ▼a(MiAaPQ)Stanfordyw528ms1740
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a306
■1001 ▼aWang, Rose Elizabeth.
■24510▼aScaling Expertise Via Language Models: with Applications to Education
■260 ▼a[Sl]▼bStanford University▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a283 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-01, Section: A.
■500 ▼aAdvisor: Demszky, Dora;Yang, Diyi.
■5021 ▼aThesis (Ph.D.)--Stanford University, 2025.
■520 ▼aAccess to expertise shapes how individuals learn, develop, and succeed across society. For example, in education, experienced teachers teach students and train novice educators through effective interactions. However, access to expertise is limited, undermining learning at scale. While language models promise to democratize access, they often mimic surface-level patterns and lack the human touch needed to support learners through challenges. In this thesis, I present a new paradigm, scalable expertise, that reimagines how we can extend human-like expertise by adapting and leveraging language models for high-stakes, real-world interactions. In Part I, I introduce a computational method to extract expert reasoning from verbalized talk aloud protocols, enabling language models to handle complex tasks. In Part II, I introduce benchmarks aligned to domain needs and, resultingly, provide early insights into the reliability of language models for intervening on real-world interactions. In Part III, I introduce real-world applications of scaling expertise, including interventions with real-time AI assistance and methods to surface interaction patterns between real teachers and students. Finally, I conclude by discussing open questions and future directions for the paradigm of scalable expertise.The work presented in this thesis has broad implications: it demonstrates how language models---when adapted, evaluated and deployed appropriately---can enhance the human touch to improve real-world behaviors and outcomes at scale. While this thesis focuses on applications to education, it establishes the foundation for scaling expertise to other high-stakes domains like healthcare and law. By addressing the challenges of methods, evaluations, and applications, this thesis advances a human-centered vision of AI enhancing human learning at scale.
■590 ▼aSchool code: 0212.
■650 4▼aFamilies & family life
■650 4▼aTutoring
■650 4▼aEducation
■690 ▼a0515
■71020▼aStanford University.
■7730 ▼tDissertations Abstracts International▼g87-01A.
■790 ▼a0212
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17358062▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
