|
JAIST Repository >
JAIST >
Theses >
Doctor of Philosophy（Information Science） >
R07) (Jun.2025 - Mar.2026 >
Please use this identifier to cite or link to this item:
https://hdl.handle.net/10119/20074
|
| Title: | 法的言語モデルにおける幻覚の検出と理解：不確実性と内部シグナルに基づく複合的手法の研究 |
| Authors: | DANG, HOANG ANH |
| Authors(alternative): | だん, ほあん あん |
| Keywords: | Hallucination Detection
Large Languge Models (LLMs)
Legal Question Answering
Internal Scoring Signals
CUAD-QA dataset |
| Issue Date: | Sep-2025 |
| Description: | Supervisor: NGUYEN, Minh Le
先端科学技術研究科
博士 |
| Title(English): | Detecting and Understanding Hallucinations in Legal Language Models: A Multi-Method Study of Uncertainty and Internal Signals |
| Authors(English): | DANG, HOANG ANH |
| Language: | eng |
| URI: | https://hdl.handle.net/10119/20074 |
| Academic Degrees and number: | 甲第1572号 |
| Degree-granting date: | 2025-09-24 |
| Degree name: | 博士 (情報科学) |
| Degree-granting institutions: | 北陸先端科学技術大学院大学 |
| Appears in Collections: | D-IS. 2025年度(R07) (Jun.2025 - Mar.2026)
|
Files in This Item:
| File |
Description |
Size | Format |
|---|
| abstract.pdf | 要旨 | 32Kb | Adobe PDF | View/Open | | paper.pdf | 本文 | 1905Kb | Adobe PDF | View/Open | | summary.pdf | 内容の要旨及び論文審査の結果の要旨 | 301Kb | Adobe PDF | View/Open |
|
All items in DSpace are protected by copyright, with all rights reserved.
|
