Newest papers

Tracing Hallucinations in LLM Reasoning for UML Class Diagram Generation: An Empirical Taxonomy and Mitigation Study

Mojdeh Rahmanian, Foutse Khomh, Ashkan Sami, Yanchao Yu · 2026 · arXiv

DOI: 10.48550/arXiv.2610.30946 · open-access PDF

Indonesian

Model Bahasa Besar (LLM) dapat menghasilkan diagram kelas UML dari persyaratan bahasa alami, tetapi halusinasi dapat membahayakan keakuratan dan kelengkapan diagram yang dihasilkan. Evaluasi yang ada sebagian besar mengevaluasi diagram akhir, tanpa memeriksa jejak pertengahan dari LLM yang mampu berpendapat, yang dapat mengungkapkan di mana halusinasi ini terjadi. Kami mengembangkan taksonomi empiris dari sebelas sub-tipe halusinasi yang diatur menjadi tiga dimensi: Prompt Faithfulness (kecilnya antara persyaratan dan argumen), Code Faithfulness (kecilnya antara berpendapat dan P yang dihasilkan

Machine translation (Meta M2M100); the English original follows.

Abstract

Large Language Models (LLMs) can generate UML class diagrams from natural language requirements, but hallucinations may compromise the correctness and completeness of the resulting diagrams. Existing evaluations mostly assess the final diagram, without examining the intermediate traces of reasoning-capable LLMs, which can reveal where these hallucinations arise. We develop an empirical taxonomy of eleven hallucination sub-types organised into three dimensions: Prompt Faithfulness (deviations between requirements and reasoning), Code Faithfulness (discrepancies between reasoning and generated P

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.30946, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange