ETF: An Entity Tracing Framework for Hallucination Detection in Code Summaries

Збережено в:

Бібліографічні деталі
Опубліковано в::	arXiv.org (Dec 18, 2024), p. n/a
Автор:	Maharaj, Kishan
Інші автори:	Munigala, Vitobha, Tamilselvam, Srikanth G, Kumar, Prince, Sen, Sayandeep, Kodeswaran, Palani, Mishra, Abhijit, Bhattacharyya, Pushpak
Опубліковано:	Cornell University Library, arXiv.org
Предмети:	Program verification (computers) Large language models Summaries Tracing Task complexity
Онлайн доступ:	Citation/Abstract Full text outside of ProQuest
Теги:	Додати тег Немає тегів, Будьте першим, хто поставить тег для цього запису!

Опис
Короткий огляд:	Recent advancements in large language models (LLMs) have significantly enhanced their ability to understand both natural language and code, driving their use in tasks like natural language-to-code (NL2Code) and code summarization. However, LLMs are prone to hallucination-outputs that stray from intended meanings. Detecting hallucinations in code summarization is especially difficult due to the complex interplay between programming and natural languages. We introduce a first-of-its-kind dataset with \(\sim\)10K samples, curated specifically for hallucination detection in code summarization. We further propose a novel Entity Tracing Framework (ETF) that a) utilizes static program analysis to identify code entities from the program and b) uses LLMs to map and verify these entities and their intents within generated code summaries. Our experimental analysis demonstrates the effectiveness of the framework, leading to a 0.73 F1 score. This approach provides an interpretable method for detecting hallucinations by grounding entities, allowing us to evaluate summary accuracy.
ISSN:	2331-8422
Джерело:	Engineering Database