Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2503.05587 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913060285841408 |
|---|---|
| author | Yang, Shiping Wu, Jie Ding, Wenbiao Wu, Ning Liang, Shining Gong, Ming Li, Hongzhi Zhang, Hengyuan Chang, Angel X. Zhang, Dongmei |
| author_facet | Yang, Shiping Wu, Jie Ding, Wenbiao Wu, Ning Liang, Shining Gong, Ming Li, Hongzhi Zhang, Hengyuan Chang, Angel X. Zhang, Dongmei |
| contents | Robustness has become a critical attribute for the deployment of RAG systems in real-world applications. Existing research focuses on robustness to explicit noise (e.g., document semantics) but overlooks implicit noise (spurious features). Moreover, previous studies on spurious features in LLMs are limited to specific types (e.g., formats) and narrow scenarios (e.g., ICL). In this work, we identify and study spurious features in the RAG paradigm, a robustness issue caused by the sensitivity of LLMs to semantic-agnostic features. We then propose a novel framework, SURE, to empirically quantify the robustness of RALMs against spurious features. Beyond providing a comprehensive taxonomy and metrics for evaluation, the framework's data synthesis pipeline facilitates training-based strategies to improve robustness. Further analysis suggests that spurious features are a widespread and challenging problem in the field of RAG. Our code is available at https://github.com/maybenotime/RAG-SpuriousFeatures . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_05587 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Quantifying and Improving the Robustness of Retrieval-Augmented Language Models Against Spurious Features in Grounding Data Yang, Shiping Wu, Jie Ding, Wenbiao Wu, Ning Liang, Shining Gong, Ming Li, Hongzhi Zhang, Hengyuan Chang, Angel X. Zhang, Dongmei Computation and Language Artificial Intelligence Machine Learning Robustness has become a critical attribute for the deployment of RAG systems in real-world applications. Existing research focuses on robustness to explicit noise (e.g., document semantics) but overlooks implicit noise (spurious features). Moreover, previous studies on spurious features in LLMs are limited to specific types (e.g., formats) and narrow scenarios (e.g., ICL). In this work, we identify and study spurious features in the RAG paradigm, a robustness issue caused by the sensitivity of LLMs to semantic-agnostic features. We then propose a novel framework, SURE, to empirically quantify the robustness of RALMs against spurious features. Beyond providing a comprehensive taxonomy and metrics for evaluation, the framework's data synthesis pipeline facilitates training-based strategies to improve robustness. Further analysis suggests that spurious features are a widespread and challenging problem in the field of RAG. Our code is available at https://github.com/maybenotime/RAG-SpuriousFeatures . |
| title | Quantifying and Improving the Robustness of Retrieval-Augmented Language Models Against Spurious Features in Grounding Data |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2503.05587 |