通過基於事實的歸因和學習拒絕來衡量和增強RAG中LLM的可信度。
Measuring and Enhancing Trustworthiness of LLMs in RAG through Grounded Attributions and Learning to Refuse
September 17, 2024
作者: Maojia Song, Shang Hong Sim, Rishabh Bhardwaj, Hai Leong Chieu, Navonil Majumder, Soujanya Poria
cs.AI
摘要
LLM 是檢索增強生成(RAG)系統中不可或缺的一部分。雖然許多研究集中於評估端到端 RAG 系統的質量,但對於 LLM 在 RAG 任務中的適用性缺乏研究。因此,我們引入了一個新的指標,Trust-Score,它提供了對於在 RAG 框架中的 LLM 的可信度的全面評估。我們展示了各種提示方法,如上下文學習,未能有效地使 LLM 適應 RAG 任務。因此,我們提出了 Trust-Align,一個用於對齊 LLM 以獲得更高 Trust-Score 的框架。與我們的方法對齊的 LLaMA-3-8b,在 ASQA(提高 10.7)、QAMPARI(提高 29.2)和 ELI5(提高 14.9)上明顯優於相同大小的開源 LLM。我們在以下位置釋出我們的代碼:https://github.com/declare-lab/trust-align。
English
LLMs are an integral part of retrieval-augmented generation (RAG) systems.
While many studies focus on evaluating the quality of end-to-end RAG systems,
there is a lack of research on understanding the appropriateness of an LLM for
the RAG task. Thus, we introduce a new metric, Trust-Score, that provides a
holistic evaluation of the trustworthiness of LLMs in an RAG framework. We show
that various prompting methods, such as in-context learning, fail to adapt LLMs
effectively to the RAG task. Thus, we propose Trust-Align, a framework to align
LLMs for higher Trust-Score. LLaMA-3-8b, aligned with our method, significantly
outperforms open-source LLMs of comparable sizes on ASQA (up 10.7), QAMPARI (up
29.2) and ELI5 (up 14.9). We release our code at:
https://github.com/declare-lab/trust-align.Summary
AI-Generated Summary