Making Retrieval-Augmented Language Models Robust to Irrelevant Context

Ranked by how close their title and abstract are to this paper’s. Select some to read them together.

Closest papers

1. Very close Another approach to the problem Not relevant

   Enhancing Robustness of Retrieval-Augmented Language Models with In-Context Learning

   Seong-Il Park, Seungwoo Choi, Na-Hyun Kim et al. · 2024 · KNOWLEDGENLP · cited 9 times · Not summarized yet

   Enhancing Robustness of Retrieval-Augmented Language Models with In-Context Learning uses MRC demonstrations to improve robustness to unanswerable and conflicting retrieved contexts.

   Abstract

   Retrieval-Augmented Language Models (RALMs) have significantly improved performance in open-domain question answering (QA) by leveraging external knowledge. However, RALMs still struggle with unanswerable queries, where the retrieved contexts do not contain the correct answer, and with conflicting information, where different sources provide contradictory answers due to imperfect retrieval. This study introduces an in-context learning-based approach to enhance the reasoning capabilities of RALMs, making them more robust in imperfect retrieval scenarios. Our method incorporates Machine Reading Comprehension (MRC) demonstrations, referred to as \textit{cases}, to boost the model’s capabilities…

2. Very close Builds on this paper Cites this paper Not relevant

   Toward Robust RALMs: Revealing the Impact of Imperfect Retrieval on Retrieval-Augmented Language Models

   Seong-Il Park, Jay-Yoon Lee · 2024 · Transactions of the Association for Computational Linguistics · cited 3 times · Not summarized yet

   Toward Robust RALMs investigates unanswerable, adversarial, and conflicting retrieval scenarios, extending analysis of how imperfect evidence affects RALM performance.

   Abstract

   Abstract Retrieval Augmented Language Models (RALMs) have gained significant attention for their ability to generate accurate answers and improve efficiency. However, RALMs are inherently vulnerable to imperfect information due to their reliance on the imperfect retriever or knowledge source. We identify three common scenarios—unanswerable, adversarial, conflicting—where retrieved document sets can confuse RALMs with plausible real-world examples. We present the first comprehensive investigation to assess how well RALMs detect and handle such problematic scenarios. Among these scenarios, to systematically examine adversarial robustness we propose a new adversarial attack method, Generative m…

3. Very close Groundwork for this paper Cited by this paper A method it uses Not relevant

   In-Context Retrieval-Augmented Language Models

   Ori Ram, Yoav Levine, Itay Dalmedigos et al. · 2023 · Computation and Language · cited 1,190 times · AI summary available

   In-Context Retrieval-Augmented Language Models conditions an unchanged language model on retrieved documents, an approach used as a foundation for retrieval-augmented systems.

   Abstract

   Retrieval-Augmented Language Modeling (RALM) methods, that condition a language model (LM) on relevant documents from a grounding corpus during generation, have been shown to significantly improve language modeling while also providing a natural source attribution mechanism. Existing RALM approaches focus on modifying the LM architecture in order to facilitate the incorporation of external information, significantly complicating deployment. This paper proposes an under-explored alternative, which we dub In-Context RALM: leaving the LM architecture unchanged and prepending grounding documents to the input. We show that in-context RALM which uses off-the-shelf general purpose retrievers provid…

4. Very close Builds on this paper Cites this paper Not relevant

   Where Does Robustness Live? Neuron-Guided Adaptation for Retrieval-Augmented Language Models

   Jae Lee, Jaemin Kim, Sumyeong Ahn et al. · 2026 · Not summarized yet

   Where Does Robustness Live? separates abstention from evidence extraction and uses neuron-guided tuning to address noisy or irrelevant contexts.

   Abstract

   Retrieval-Augmented Language Models (RALMs) have shown strong potential in knowledge-intensive tasks, yet they remain vulnerable when retrieved contexts are noisy or irrelevant. Robustness against such contexts requires two distinct capabilities: abstention when contexts are uninformative, and selective extraction when relevant evidence is buried in noise. Yet existing methods face two key limitations: they do not train separately for these two capabilities, and they adapt the model at a coarse layer- or module-level granularity, overlooking that only a small subset of neurons is strongly activated for a given input. We propose NeuRIT, a Neuron-guided Robust Instruction-Tuning framework buil…

5. Very close Another approach to the problem Not relevant

   Improving Retrieval Augmented Language Model with Self-Reasoning

   Yuan Xia, Jingbo Zhou, Zhenhui Shi et al. · 2024 · Computation and Language · cited 53 times · AI summary available

   Improving Retrieval Augmented Language Model with Self-Reasoning targets performance losses from irrelevant documents using a self-reasoning framework.

   Abstract

   The Retrieval-Augmented Language Model (RALM) has shown remarkable performance on knowledge-intensive tasks by incorporating external knowledge during inference, which mitigates the factual hallucinations inherited in large language models (LLMs). Despite these advancements, challenges persist in the implementation of RALMs, particularly concerning their reliability and traceability. To be specific, the irrelevant document retrieval may result in unhelpful response generation or even deteriorate the performance of LLMs, while the lack of proper citations in generated outputs complicates efforts to verify the trustworthiness of the models. To this end, we propose a novel self-reasoning framew…

6. Very close Another approach to the problem Not relevant

   Chain-of-Note: Enhancing Robustness in Retrieval-Augmented Language Models

   Wenhao Yu, Hongming Zhang, Xiaoman Pan et al. · 2023 · Computation and Language · cited 263 times · AI summary available

   Chain-of-Note addresses misleading retrieved data by helping RALMs assess available knowledge and avoid overlooking what they already know.

   Abstract

   Retrieval-augmented language models (RALMs) represent a substantial advancement in the capabilities of large language models, notably in reducing factual hallucination by leveraging external knowledge sources. However, the reliability of the retrieved information is not always guaranteed. The retrieval of irrelevant data can lead to misguided responses, and potentially causing the model to overlook its inherent knowledge, even when it possesses adequate information to address the query. Moreover, standard RALMs often struggle to assess whether they possess adequate knowledge, both intrinsic and retrieved, to provide an accurate answer. In situations where knowledge is lacking, these systems …

7. Very close Same question Not relevant

   Unraveling and Mitigating Retriever Inconsistencies in Retrieval-Augmented Large Language Models

   Ming-Da Li, Xin-Yu Li, Yifan Chen et al. · 2024 · Annual Meeting of the Association for Computational Linguistics · cited 2 times · Not summarized yet

   Unraveling and Mitigating Retriever Inconsistencies studies why retrieval can reduce RALM accuracy, including reader degeneration and differences among knowledge sources.

   Abstract

   Although Retrieval-Augmented Large Language Models (RALMs) demonstrate their superiority in terms of factuality, they do not consistently outperform the original retrieval-free Language Models (LMs). Our experiments reveal that this example-level performance inconsistency exists not only between retrieval-augmented and retrieval-free LM but also among different retrievers. To understand this phenomenon, we investigate the degeneration behavior of RALMs and theoretically decompose it into four categories. Further analysis based on our decomposition reveals that the innate difference in knowledge sources and the unpredictable degeneration of the reader model contribute most to the inconsistenc…

8. Very close Same question Cites this paper Not relevant

   Evaluating the Retrieval Robustness of Large Language Models

   Shuyang Cao, Karthik Radhakrishnan, David S. Rosenberg et al. · 2025 · arXiv.org · preprint, not peer reviewed · cited 4 times · Not summarized yet

   Evaluating the Retrieval Robustness of Large Language Models tests when retrieval harms performance and how document quantity and order affect RAG results.

   Abstract

   Retrieval-augmented generation (RAG) generally enhances large language models'(LLMs) ability to solve knowledge-intensive tasks. But RAG could also lead to performance degradation due to imperfect retrieval and the model's limited ability to leverage retrieved content. In this work, we evaluate the robustness of LLMs in practical RAG setups (henceforth retrieval robustness). We focus on three research questions: (1) whether RAG is always better than non-RAG; (2) whether more retrieved documents always lead to better performance; and (3) whether document order impacts results. To facilitate this study, we establish a benchmark of 1,891 samples spanning five datasets across three task categori…

Prior work it builds on

1. Close Groundwork for this paper Used in depth by this paper A method it uses Not relevant

   ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT

   O. Khattab, M. Zaharia · 2020 · Annual International ACM SIGIR Conference on Research and Development in Information Retrieval · cited 2,804 times · AI summary available

   ColBERT presents an efficient passage-ranking method that can supply retrieved documents to retrieval-augmented language models.

   Abstract

   Recent progress in Natural Language Understanding (NLU) is driving fast-paced advances in Information Retrieval (IR), largely owed to fine-tuning deep language models (LMs) for document ranking. While remarkably effective, the ranking models based on these LMs increase computational cost by orders of magnitude over prior approaches, particularly as they must feed each query-document pair through a massive neural network to compute a single relevance score. To tackle this, we present ColBERT, a novel ranking model that adapts deep LMs (in particular, BERT) for efficient retrieval. ColBERT introduces a late interaction architecture that independently encodes the query and the document using BE…

2. Very close Broader context Not relevant

   Long-range Language Modeling with Self-retrieval

   Ohad Rubin, Jonathan Berant · 2023 · Computation and Language · cited 34 times · AI summary available

   Long-range Language Modeling with Self-retrieval trains a retrieval-augmented transformer to retrieve earlier text chunks for long-document language modeling.

   Abstract

   Retrieval-augmented language models (LMs) have received much attention recently. However, typically the retriever is not trained jointly as a native component of the LM, but added to an already-pretrained LM, which limits the ability of the LM and the retriever to adapt to one another. In this work, we propose the Retrieval-Pretrained Transformer (RPT), an architecture and training procedure for jointly training a retrieval-augmented LM from scratch for the task of modeling long texts. Given a recently generated text chunk in a long document, the LM computes query representations, which are then used to retrieve earlier chunks in the document, located potentially tens of thousands of tokens …

3. Close Broader context A method it uses Not relevant

   Active Retrieval Augmented Generation

   Zhengbao Jiang, Frank F. Xu, Luyu Gao et al. · 2023 · Computation and Language · cited 953 times · AI summary available

   Active Retrieval Augmented Generation retrieves information repeatedly during long-form generation rather than using a single retrieval step.

   Abstract

   Despite the remarkable ability of large language models (LMs) to comprehend and generate language, they have a tendency to hallucinate and create factually inaccurate output. Augmenting LMs by retrieving information from external knowledge resources is one promising solution. Most existing retrieval-augmented LMs employ a retrieve-and-generate setup that only retrieves information once based on the input. This is limiting, however, in more general scenarios involving generation of long texts, where continually gathering information throughout the generation process is essential. There have been some past efforts to retrieve information multiple times while generating outputs, which mostly re…

4. Close Groundwork for this paper Not relevant

   REALM: Retrieval-Augmented Language Model Pre-Training

   Kelvin Guu, Kenton Lee, Zora Tung et al. · 2020 · International Conference on Machine Learning · cited 3,602 times · Not summarized yet

   REALM introduces a latent knowledge retriever in language-model pretraining, fine-tuning, and inference.

   Abstract

   Language model pre-training has been shown to capture a surprising amount of world knowledge, crucial for NLP tasks such as question answering. However, this knowledge is stored implicitly in the parameters of a neural network, requiring ever-larger networks to cover more facts. To capture knowledge in a more modular and interpretable way, we augment language model pre-training with a latent knowledge retriever, which allows the model to retrieve and attend over documents from a large corpus such as Wikipedia, used during pre-training, fine-tuning and inference. For the first time, we show how to pre-train such a knowledge retriever in an unsupervised manner, using masked language modeling a…

5. Close Groundwork for this paper Not relevant

   Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

   Patrick Lewis, Ethan Perez, Aleksandra Piktus et al. · 2020 · Computation and Language · cited 20,501 times · AI summary available

   Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks uses external non-parametric memory to support language models on knowledge-intensive tasks.

   Abstract

   Large pre-trained language models have been shown to store factual knowledge in their parameters, and achieve state-of-the-art results when fine-tuned on downstream NLP tasks. However, their ability to access and precisely manipulate knowledge is still limited, and hence on knowledge-intensive tasks, their performance lags behind task-specific architectures. Additionally, providing provenance for their decisions and updating their world knowledge remain open research problems. Pre-trained models with a differentiable access mechanism to explicit non-parametric memory can overcome this issue, but have so far been only investigated for extractive downstream tasks. We explore a general-purpose …

Papers that cite it

1. Very close Builds on this paper Uses this paper in depth Not relevant

   Influence Guided Context Selection for Effective Retrieval-Augmented Generation

   Jia-Le Deng, Yan-Yan Shen, Ziyuan Pei et al. · 2025 · Neural Information Processing Systems · cited 11 times · Not summarized yet

   Influence Guided Context Selection assesses retrieved-context quality using the query, context list, and generator to address noisy or irrelevant evidence.

   Abstract

   Retrieval-Augmented Generation (RAG) addresses large language model (LLM) hallucinations by grounding responses in external knowledge, but its effectiveness is compromised by poor-quality retrieved contexts containing irrelevant or noisy information. While existing approaches attempt to improve performance through context selection based on predefined context quality assessment metrics, they show limited gains over standard RAG. We attribute this limitation to their failure in holistically utilizing available information (query, context list, and generator) for comprehensive quality assessment. Inspired by recent advances in data selection, we reconceptualize context quality assessment as an…

2. Close Broader context Uses this paper in depth Not relevant

   Stable-RAG: Mitigating Retrieval-Permutation-Induced Hallucinations in Retrieval-Augmented Generation

   Qianchi Zhang, Hainan Zhang, Liang Pang et al. · 2026 · Annual Meeting of the Association for Computational Linguistics · cited 8 times · Not summarized yet

   Stable-RAG studies how permutations of retrieved documents affect answers, focusing on order sensitivity rather than irrelevant-context robustness.

   Abstract

   Retrieval-Augmented Generation (RAG) has become a key paradigm for reducing factual hallucinations in Large Language Models (LLMs), yet little is known about how the order of retrieved documents affects model behavior. We empirically show that under a Top-5 retrieval setting with the gold document included, LLM answers vary substantially across permutations of the retrieved set, even when the gold document is fixed in the first position. This reveals a previously underexplored sensitivity to retrieval permutations. Although existing robust RAG methods focus primarily on enhancing LLM robustness to low-quality retrieval and mitigating positional bias to distribute attention fairly over long c…

3. Very close Broader context Not relevant

   Investigating the Robustness of Retrieval-Augmented Generation at the Query Level

   Sezen Perçin, Xin Su, Syed Sha Qutub et al. · 2025 · arXiv.org · preprint, not peer reviewed · cited 14 times · Not summarized yet

   Investigating the Robustness of Retrieval-Augmented Generation at the Query Level examines how query perturbations affect retrieval and downstream RAG performance.

   Abstract

   Large language models (LLMs) are very costly and inefficient to update with new information. To address this limitation, retrieval-augmented generation (RAG) has been proposed as a solution that dynamically incorporates external knowledge during inference, improving factual consistency and reducing hallucinations. Despite its promise, RAG systems face practical challenges-most notably, a strong dependence on the quality of the input query for accurate retrieval. In this paper, we investigate the sensitivity of different components in the RAG pipeline to various types of query perturbations. Our analysis reveals that the performance of commonly used retrievers can degrade significantly even u…

4. Close Builds on this paper Uses this paper in depth Not relevant

   The Distracting Effect: Understanding Irrelevant Passages in RAG

   Chen Amiraz, Florin Cuconasu, Simone Filice et al. · 2025 · Annual Meeting of the Association for Computational Linguistics · cited 59 times · Not summarized yet

   The Distracting Effect quantifies how irrelevant passages mislead RAG models and fine-tunes them with selected distractors to improve accuracy.

   Abstract

   A well-known issue with Retrieval Augmented Generation (RAG) is that retrieved passages that are irrelevant to the query sometimes distract the answer-generating LLM, causing it to provide an incorrect response. In this paper, we shed light on this core issue and formulate the distracting effect of a passage w.r.t. a query (and an LLM). We provide a quantifiable measure of the distracting effect of a passage and demonstrate its robustness across LLMs. Our research introduces novel methods for identifying and using hard distracting passages to improve RAG systems. By fine-tuning LLMs with these carefully selected distracting passages, we achieve up to a 7.5% increase in answering accuracy com…

5. Very close Broader context Not relevant

   Improving Context Fidelity via Native Retrieval-Augmented Reasoning

   Suyuchen Wang, Jin-Lin Wang, Xin-Yu Wang et al. · 2025 · Conference on Empirical Methods in Natural Language Processing · cited 8 times · Not summarized yet

   Improving Context Fidelity via Native Retrieval-Augmented Reasoning trains models to integrate in-context evidence during reasoning and answer generation.

   Abstract

   Large language models (LLMs) often struggle with context fidelity, producing inconsistent answers when responding to questions based on provided information. Existing approaches either rely on expensive supervised fine-tuning to generate evidence post-answer or train models to perform web searches without necessarily improving utilization of the given context. We propose CARE, a novel native retrieval-augmented reasoning framework that teaches LLMs to explicitly integrate in-context evidence within their reasoning process with the model's own retrieval capabilities. Our method requires limited labeled evidence data while significantly enhancing both retrieval accuracy and answer generation p…