In-Context Retrieval-Augmented Language Models
Open accessThe DOI matches its Crossref record; 53 of the 60 references it cites were found in a registry or an index; 9 of the 26 sources its claims cite were checked for support.
Abstract
Abstract Retrieval-Augmented Language Modeling (RALM) methods, which condition a language model (LM) on relevant documents from a grounding corpus during generation, were shown to significantly improve language modeling performance. In addition, they can mitigate the problem of factually inaccurate text generation and provide 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 considers a simple alternative, which we dub In-Context RALM: leaving the LM architecture unchanged and prepending grounding documents to the input, without any further training of the LM. We show that In-Context RALM that builds on off-the-shelf general purpose retrievers provides surprisingly large LM gains across model sizes and diverse corpora. We also demonstrate that the document retrieval and ranking mechanism can be specialized to the RALM setting to further boost performance. We conclude that In-Context RALM has considerable potential to increase the prevalence of LM grounding, particularly in settings where a pretrained LM must be used without modification or even via API access.1
Overview
CoChat Summary
TL;DR
In-Context RALM prepends retrieved documents to a frozen LM’s input without architecture changes or training, and off-the-shelf retrievers yield large perplexity gains across model sizes and corpora. Tailoring document ranking via zero-shot or trained rerankers further improves performance, including on open-domain QA.
Key findings
- With BM25 retrieval (`s = 4`, `ℓ = 32`), In-Context RALM improved LM perplexity enough to match a 2–3× larger model; e.g., 6.7B OPT matched 66B OPT, and 345M GPT-2 + BM25 outperformed 762M GPT-2.
- BM25 outperformed dense retrievers (Contriever, Spider, BERT mean pooling); optimal query length was `ℓ = 32` for BM25 vs. `ℓ = 64` for dense retrievers, and more frequent retrieval (smaller stride) improved performance.
- Predictive reranking improved WikiText-103 word perplexity for GPT-2 110M from 29.6 to 26.8 and GPT-2 1.5B from 16.6 to 15.4; zero-shot reranking also improved over taking the top BM25 document.
- On ODQA with two DPR documents, LLaMA-13B improved NQ exact match from 12.0% to 31.0% and TriviaQA from 54.8% to 60.1%; LLaMA-33B improved NQ from 13.7% to 32.3% and TriviaQA from 58.3% to 62.7%.
Methods
- Evaluated on five LM datasets (WikiText-103, RealNews, ArXiv, Stack Exchange, FreeLaw) and two ODQA datasets (Natural Questions, TriviaQA), using GPT-2 (110M–1.5B), GPT-Neo/GPT-J (1.3B–6B), OPT (125M–66B), and LLaMA (7B–33B) models with max sequence length 1,024.
- Retrieval corpora: Wikipedia (Dec. 20, 2018) for WikiText-103/ODQA with 120 WikiText dev/test articles removed; training data for other datasets; non-overlapping 100-word passages truncated at 256 tokens. Retrievers: BM25, BERT mean pooling, Contriever, Spider; DPR for ODQA.
- Main analyses measured perplexity on LM datasets and exact match on ODQA; reranking used top-16 BM25 documents, with zero-shot LM reranking and a RoBERTa-base predictive reranker trained for 10,000 steps (peak LR 1e-5, batch 32, 300,000 WikiText-103 examples).
Limitations
- Stated: the paper considers only prepending a single external document; adding more documents could drive further gains. It retrieves every fixed interval of `s` tokens, and future work could retrieve more sparsely (e.g., when a specialized model predicts retrieval is needed) for latency/cost gains.
- Stated: predictive reranking still leaves room compared with top-16 BM25 oracle results; oracle could improve with `k > 16` or stronger RALM-dedicated retrievers.
- Evident: main LM experiments use one document and max sequence length 1,024 despite some models supporting 2,048; the predictive reranker assumes availability of training data from the target corpus and was trained on WikiText-103.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures










Verification
Claims the paper makes
5 Valid · 4 Not valid · 9 Not checked · 43 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Their results are based on training a dedicated retriever for language modeling.”
§ In-Context Retrieval-Augmented Language Models > 1 Introduction
What the sources say
- Partly supportedcites Shi et al., 2023 matched to “REPLUG: Retrieval-augmented black-box language models” (2023) sourceExperimental — accuracy not yet measured
Section 4 describes training a retriever with language-model supervision for REPLUG LSR, and Section 5.2 gives its training procedure. But the ‘Our model’ results also include REPLUG without that retriever training, so training a dedicated retriever underlies only part of the reported results. The indexed passages provide no page numbers. (LLM, conf=0.88)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“More importantly, they tend to produce factual inaccuracies and errors (Lin et al., 2022;Maynez et al., 2020;Huang et al., 2020).”
§ In-Context Retrieval-Augmented Language Models > 1 Introduction
What the sources say
- Supportedcites Lin et al., 2022 matched to “TruthfulQA: Measuring How Models Mimic Human Falsehoods” (2022) source
The Introduction (p. 1) explicitly says that recent language models have a tendency to generate false statements, ranging from subtle inaccuracies to hallucinations. The Discussion (p. 8) also reports poor performance by the baseline models on truthfulness questions. (LLM, conf=0.94)
- Partly supportedcites Maynez et al., 2020 matched to “On Faithfulness and Factuality in Abstractive Summarization” (2020) sourceExperimental — accuracy not yet measured
Section 5.3 reports that over 90% of hallucinations from the BERTS2S summarization model are erroneous, and the Abstract reports substantial hallucinated content across the evaluated summarization systems. This supports factual errors in those systems, but the claim extends the finding to language models generally, which the paper did not evaluate. (LLM, conf=0.83)
- Partly supportedcites Huang et al., 2020 matched to “Challenges in building intelligent open-domain dialog systems” (2020) sourceExperimental — accuracy not yet measured
Section 2.3 (p. 10) says generation-based dialog methods often produce undesirable responses that are ungrammatical or irrelevant, supporting the general point about errors. Section 3.2 (p. 12) stresses grounding responses in real-world facts, but the passages do not establish that language models tend to produce factual inaccuracies specifically; the paper discusses open-domain dialog systems rather than LMs generally. (LLM, conf=0.78)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Existing RALM approaches focus on modifying the LM architecture in order to facilitate the incorporation of external information, significantly complicating deployment.”
- Also stated in: § 1 Introduction · plain
“Leading RALM systems introduced recently tend to be focused on altering the language model architecture (Khandelwal et al., 2020;Borgeaud et al., 2022;Zhong et al., 2022;Levine et al., 2022c;Li et al., 2022).”
§ In-Context Retrieval-Augmented Language Models > Abstract
What the sources say
- Not matchedcites Khandelwal et al., 2020
An author–year citation the paper’s reference list doesn’t settle. We don’t search these by name.
- Not matchedcites Borgeaud et al., 2022
An author–year citation the paper’s reference list doesn’t settle. We don’t search these by name.
- Partly supportedcites Zhong et al., 2022 matched to “Training language models with memory augmentation” (2022) sourceExperimental — accuracy not yet measured
The Introduction and §2.2 Memory Augmentation say that several earlier approaches modify Transformer architectures to incorporate memory from previous segments. They do not establish that existing retrieval-augmented language modeling approaches generally focus on architecture changes, or that those changes significantly complicate deployment. (LLM, conf=0.84)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 1 Introduction · plain
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“All prior work involves training the LM.”
§ In-Context Retrieval-Augmented Language Models > Nearest Neighbor Language Models
What the sources say
- Not matchedcites Lewis et al., 2020
More than one paper fits this citation.
- Could not verifycites Izacard and Grave, 2021 matched to “Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering” (2021) sourceExperimental — accuracy not yet measured
The Related work section (p. 2) describes earlier retrieval-augmented QA systems but does not say whether all of them train the language model. The Experiments section (p. 4) says this paper fine-tunes its own pretrained T5 model; that does not establish the claim about all prior work. I searched for discussion of training or freezing the language model in earlier retrieve-and-read systems. (LLM, conf=0.72)
- Not matchedcites Borgeaud et al., 2022
An author–year citation the paper’s reference list doesn’t settle. We don’t search these by name.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“While they showed significant gains from kNN-LM, the approach requires storing the representations for each token in the corpus, an expensive requirement even for a small corpus like Wikipedia.”
§ In-Context Retrieval-Augmented Language Models > Nearest Neighbor Language Models
What the sources say
- Partly supportedcites Zhong et al., 2022 matched to “Training language models with memory augmentation” (2022) sourceExperimental — accuracy not yet measured
Section 2.2 defines external memory as context–token pairs from a corpus and says that accessing a memory of roughly 10^8 entries is computationally expensive. The Introduction identifies kNN-LM as using external memory, and Table 2 notes large SSD storage for one retrieval setup. The paper tests on WIKITEXT-103, but these passages do not characterize Wikipedia itself as a small corpus or establish the claimed storage cost for it. (LLM, conf=0.82)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
42 more claims
Create a free account to see all 61 claims and to verify the 3 references still open.
- Probably not a substantial claim. 9
- The quote only partly states this claim. 3
- It may state two separate claims as one. 4
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“or when language models are augmented with external information (relevant documents from a search engine, database query results, etc.; Petroni et al., 2020;Ram et al., 2023;Shi et al., 2023;Mallen et al., 2023;Schick et al., 2023, inter alia).”
From Lost in the Middle: How Language Models Use Long Contexts (2024) · cites this paper as Ram et al., 2023
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In contrast to fine-tuning, RAG does not seek to adjust the internal parameters (or internal memory) of an LLM.”
From ChatGPT, extended: large language models and the extended mind (2025) · cites this paper as Ram et al., 2023
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“the task of retrieving relevant and precise information from extensive corpora can be computationally demanding and resourceintensive, necessitating continuous updating and modification of knowledge databases to keep pace with the latest information (Chen et al. 2023; Siriwardhana et al. 2023; Ram et al. 2023; Cheng et al. 2024).”
From Large Legal Fictions: Profiling Legal Hallucinations in Large Language Models (2024) · cites this paper as Ram et al. 2023
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Memory-augmented methods extend this concept by retrieving user-specific information from past interactions. This addresses the challenge of a limited context window and enables the integration of past information into the input prompt [25, 29, 45].”
From CAIM: Development and Evaluation of a Cognitive AI Memory Framework for Long-Term Interaction with Intelligent Agents (2026) · cites this paper as [45]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Despite their capabilities, LLMs still have shortcomings, such as a limited context window, which restricts the amount of information that can be processed and thus the ability to understand the ever-changing environment [58, 60] and maintain effective long-term interactions [45, 50].”
From CAIM: Development and Evaluation of a Cognitive AI Memory Framework for Long-Term Interaction with Intelligent Agents (2026) · cites this paper as [45]
Experimental — accuracy not yet measured
9 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ In-Context Retrieval-Augmented Language Models > 8 Discussion
“this paper considers only the case of prepending a single external document to the context”
§ In-Context Retrieval-Augmented Language Models > 7 In-Context RALM for Open-Domain Question Answering
“In the open-book setting, we include the top two documents returned by DPR.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ In-Context Retrieval-Augmented Language Models > Nearest Neighbor Language Models
“All prior work involves training the LM.”
§ In-Context Retrieval-Augmented Language Models > Nearest Neighbor Language Models
“Levine et al. (2022a) showed competitive open domain question-answering performance by prompt-tuning a frozen LM as a reader.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ In-Context Retrieval-Augmented Language Models > Model
“Table 3 shows that a small LM (GPT-2 117M) can be used to rerank the documents for all larger GPT-2 models”
§ In-Context Retrieval-Augmented Language Models > Model
“GPT-2 345M (M) GPT-2 110M (S) 20.8 12.1”
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019) · arXiv · doi:10.48550/arxiv.1907.11692 · Cited by 16,515
Advances in neural information processing systems 7
(1996) · Computers & Mathematics with Applications · doi:10.1016/s0898-1221(96)90463-0 · Cited by 14,281
Advances in Neural Information Processing Systems 29
Teymur, O., Zygalakis, K., & Calderhead, B. (2016) · Cited by 13,887
Transformers: State-of-the-Art Natural Language Processing
Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., Cistac, P., Rault, T., Louf, R., Funtowicz, M., et al. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations · doi:10.18653/v1/2020.emnlp-demos.6 · Cited by 8,741
LLaMA: Open and Efficient Foundation Language Models
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.-A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.13971 · Cited by 4,602
Billion-scale similarity search with GPUs
Johnson, J., Douze, M., & Jegou, H. (2021) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2019.2921572 · Cited by 4,228 · cited 1×
“For dense retrieval, we applied exact search using FAISS (Johnson et al., 2021).”
MethodBillion-scale similarity search with GPUs
Johnson, J., Douze, M., & Jégou, H. (2017) · arXiv · doi:10.48550/arxiv.1702.08734 · Cited by 4,140 · cited 1×
“For dense retrieval, we applied exact search using FAISS (Johnson et al., 2021).”
Method
Research works printed but not linked: 1
- 17. Yoav Levine, Ori Ram, Daniel Jannai, Barak Lenz, Shai Shalev-Shwartz, Amnon Shashua, Kevin Leyton-Brown, and Yoav Shoham. 2022b. Huge frozen language models as readers for open-domain question answering. In ICML 2022 Workshop on Knowledge Retrieval and Language Models.
Cited by
Items 101–200 of 454
FT2Ra: A Fine-Tuning-Inspired Approach to Retrieval-Augmented Code Completion
Guo, Q., Li, X., Xie, X., Liu, S., Tang, Z. J., Feng, R., Wang, J., Ge, J., & Bu, L. (2024) · ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA) · doi:10.1145/3650212.3652130 · Cited by 28
PathOCL: Path-Based Prompt Augmentation for OCL Generation with GPT-4
Abukhalaf, S., Hamdaqa, M., & Khomh, F. (2024) · IEEE/ACM International Conference on AI Foundation Models and Software Engineering (FORGE) · doi:10.1145/3650105.3652290 · Cited by 30
Retrieval Augmented Recipe Generation
Liu, G., Yin, H., Zhu, B., Chen, J., Ngo, C., & Jiang, Y. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00244 · Cited by 33
“On the one hand, we propose Stochastic Diversi-fied Retrieval Augmentation (SDRA) to provide a rich and diverse set of retrieved recipes as references, which could provide relevant knowledge for the generation to reduce hallucination [24, 29, 52].”
Wei, S., Hu, A., Liang, Y., Yang, J., Yu, L., Li, W., Yang, B., & Qiu, J. (2025) · Radiation Oncology · doi:10.1186/s13014-025-02660-5 · Cited by 28
“Additionally, investigating more effective prompts, incorporating Retrieval Augmented Generation (RAG) technology [52–54], or experimenting with Controlled Text Generation (CTG) methods [55, 56] could also be explored to enhance the reliability of utilizing smaller LLMs to adjust plan optimization…”
Toward Holistic Evaluation of Recommender Systems Powered by Generative Models
Deldjoo, Y., Mehta, N., Sathiamoorthy, M., Zhang, S., Castells, P., & McAuley, J. (2025) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3726302.3730354 · Cited by 27
Harvel, N., Haiek, F. B., Ankolekar, A., & Brunner, D. J. (2024) · Proceedings of the AAAI Symposium Series · doi:10.1609/aaaiss.v3i1.31191 · Cited by 21
“Prior work (Xu et al. 2023) has demonstrated the value of well-chosen context, as in retrieval-augmented generation (RAG) (Ram et al. 2023), as being more effective than simply increasing the context window of LLMs for long context tasks such as query-based summarization and document…”
BackgroundMethodInfluentialCPR: Retrieval Augmented Generation for Copyright Protection
Golatkar, A., Achille, A., Zancato, L., Wang, Y., Swaminathan, A. D., & Soatto, S. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.01176 · Cited by 89
“Retrieval Augmented Generation Retrival Augmented Generation (RAG) methods have been successfully applied to large language models (LLMs) [24, 31, 42, 45, 54].”
MethodEmbodied AI-Guided Interactive Digital Teachers for Education
Zhao, Z., Yin, Z., Sun, J., & Hui, P. (2024) · doi:10.1145/3680533.3697070 · Cited by 26
“This can be achieved through fine-tuning the LLM or even using a training-free approach [Ram et al. 2023], making it an effective and efficient solution for improving LLM performance.”
Knowing You Don't Know: Learning When to Continue Search in Multi-round RAG through Self-Practicing
Yang, D., Zeng, L., Rao, J., & Zhang, Y. (2025) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3726302.3730018 · Cited by 41
Hannah, G., Sousa, R. T., Dasoulas, I., & d’Amato, C. (2024) · Journal of Web Semantics · doi:10.1016/j.websem.2024.100843 · Cited by 24
Evaluating ChatGPT on Korea's BIM Expertise Exam and improving its performance through RAG
Yu, Y., Kim, S. K., Lee, W., & Koo, B. (2025) · Journal of Computational Design and Engineering · doi:10.1093/jcde/qwaf033 · Cited by 20
REMED: Retrieval-Augmented Medical Document Query Responding with Embedding Fine-Tuning
Pang, T., Tan, K., Yao, Y., Liu, X., Meng, F., Fan, C., & Zhang, X. (2024) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn60899.2024.10651011 · Cited by 17
“Recent studies have shown that by retrieving and enhancing the model’s input with similar vocabulary or text fragments to the current task, its generalization ability and handling of unknown data can significantly improved [3], [4], [7], [8].”
BackgroundPasaribu, J., Yudistira, N., & Mahmudy, W. F. (2024) · IEEE Access · doi:10.1109/access.2024.3518205 · Cited by 33
“These methods not only boost performance but also help mitigate factually inaccurate text generation and provide a natural source attribution mechanism [29].”
Structured knowledge-based causal discovery: Agentic streams of thought
Meier, S., Raut, P. N., Mahr, F., Thielen, N., Franke, J. E., & Risch, F. (2025) · Information Processing & Management · doi:10.1016/j.ipm.2025.104202 · Cited by 16
Patterson, B. W., Hekman, D. J., Liao, F., Hamedani, A. G., Shah, M. N., & Afshar, M. (2024) · JAMIA Open · doi:10.1093/jamiaopen/ooae039 · Cited by 15
“25 One limitation of current LLMs, however, is the constraint on the number of tokens that can be processed at once.”
BackgroundDr.ICL: Demonstration-Retrieved In-context Learning
Luo, M., Xu, X., Dai, Z., Pasupat, P., Kazemi, M., Baral, C. R., Imbrasaitė, V., & Zhao, V. Y. (2024) · Data Intelligence · doi:10.3724/2096-7004.di.2024.0012 · Cited by 109
Guo, C., Tian, Z., Tang, J., Li, S., & Wang, T. (2024) · Information Processing & Management · doi:10.1016/j.ipm.2024.103978 · Cited by 19
Improving LLM Reliability with RAG in Religious Question-Answering: MufassirQAS
Alan, A. Y., Karaarslan, E., & Aydın, Ö. (2025) · Turkish Journal of Engineering · doi:10.31127/tuje.1624773 · Cited by 16
“There are studies in the literature where RAG is used in NLP tasks [22-25].”
Knowledge-Centered Dual-Process Reasoning for Math Word Problems With Large Language Models
Liu, J., Huang, Z., Liu, Q., Ma, Z., Zhai, C., & Chen, E. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3556367 · Cited by 19
“Future advancements in this technique focused either on designing different retrieval methods, such as using sparse [52], [53] or dense [54], [55] retrieval, or on employing various strategies to integrate the retrieved content into the model’s generation process, such as integrating at LLMs’ input…”
Max–Min semantic chunking of documents for RAG application
Kiss, C., Nagy, M., & Szilágyi, P. (2025) · Discover Computing · doi:10.1007/s10791-025-09638-7 · Cited by 18
“The simplest approach splits text into fixed-sized chunks, typically based on a predefined word (e.g., 100-word chunks, [1, 10]) or token count (e.g, 288-token chunks, [11]).”
Levich, S., & Knust, L. (2025) · International Journal of Accounting Information Systems · doi:10.1016/j.accinf.2025.100750 · Cited by 17
LMCap: Few-shot Multilingual Image Captioning by Retrieval Augmented Language Model Prompting
Ramos, R., Martins, B., & Elliott, D. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.104 · Cited by 34
CRUSH4SQL: Collective Retrieval Using Schema Hallucination For Text2SQL
Kothyari, M., Dhingra, D., Sarawagi, S., & Chakrabarti, S. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.868 · Cited by 49
Alleviating Hallucination in Large Vision-Language Models with Active Retrieval Augmentation
Qu, X., Chen, Q., Wei, W., Sun, J., Liu, D., & Dong, J. (2025) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3742434 · Cited by 45
HeterRAG: Heterogeneous Processing-in-Memory Acceleration for Retrieval-augmented Generation
Liu, C., Liu, H., Chen, D., Huang, Y., Zhang, Y., Xiao, W., Liao, X., & Jin, H. (2025) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1145/3695053.3731089 · Cited by 26
“To address this limitation, the Retrieval-augmented Generation (RAG) method has emerged [7, 59, 68, 74].”
In-Storage Acceleration of Retrieval Augmented Generation as a Service
Mahapatra, R., Santhanam, H., Priebe, C., Xu, H., & Esmaeilzadeh, H. (2025) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1145/3695053.3731032 · Cited by 24
Ahn, J., Kang, B. G., Chang, M. Y., & Yoon, S. (2025) · Clinical and Experimental Otorhinolaryngology · doi:10.21053/ceo.2025-00121 · Cited by 10
“To address these issues, various grounding techniques have been proposed that leverage external resources, including documents [37], images [38], and videos [39].”
Park, G., Park, J., & Lee, H. (2025) · Applied Sciences · doi:10.3390/app15147800 · Cited by 12
RETTA: Retrieval-enhanced test-time adaptation for zero-shot video captioning
Ma, Y., Qing, L., Li, G., Qi, Y., Beheshti, A., Sheng, Q. Z., & Huang, Q. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.112170 · Cited by 18
“This is because traditional RAG methods [13, 14] typically use the question as a query to retrieve relevant textual information as a prefix, which e ff ectively guides the model to generate accurate responses.”
Wang, Y., Liu, Q., Jiang, Z., Wang, T., Jiao, J., Chu, H., Gao, B., & Chen, H. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00369 · Cited by 25
Li, S., Chen, J., Yuan, S., Wu, X., Yang, H., Tao, S., & Xiao, Y. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i17.29817 · Cited by 77
RepoMinCoder: Improving Repository-Level Code Generation Based on Information Loss Screening
Li, Y., Shi, E., Zheng, D., Duan, K., Chen, J., & Wang, Y. (2024) · Asia-Pacific Symposium on Internetware · doi:10.1145/3671016.3674819 · Cited by 25
Goldfish: Vision-Language Understanding of Arbitrarily Long Videos
Ataallah, K., Shen, X., Abdelrahman, E., Sleiman, E., Zhuge, M., Ding, J., Zhu, D., Schmidhuber, J., & Elhoseiny, M. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73397-0_15 · Cited by 68
Soekamto, Y., Limanjaya, L. C., Purwanto, Y. K., & Kang, D.-K. (2025) · IEEE Access · doi:10.1109/access.2025.3535618 · Cited by 7
“Advancements in RAG techniques have incorporated multi-hop retrieval [36], [37], [38] and cross-attention mechanisms [39], [40], enabling the model to retrieve and synthesize information from multiple sources for more complex queries.”
Gotoh, R. (2025) · Energy Reports · doi:10.1016/j.egyr.2025.02.016 · Cited by 11
Jiang, P., Han, J., & Ahmed‐Kristensen, S. (2025) · Sustainable Futures · doi:10.1016/j.sftr.2025.101094 · Cited by 18
Orchestrating Agents and Data for Enterprise: A Blueprint Architecture for Compound AI
Kandogan, E., Bhutani, N., Zhang, D., Chen, R. L., Gurajada, S., & Junior, E. R. H. (2025) · IEEE International Conference on Data Engineering Workshops (ICDEW) · doi:10.1109/icdew67478.2025.00007 · Cited by 31
“Approaches range from utilizing a single data source [18] to heterogeneous, multi-modal data sources [6], with the execution being a single-step [37], [53] or interleaved with reasoning [52].”
Improving Text Embedding Models with Positive-aware Hard-negative Mining
Moreira, G. D. S. P., Osmulski, R., Xu, M., Ak, R., Schifferer, B., & Oldridge, E. (2025) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3746252.3761254 · Cited by 22
A Multi-Source Retrieval Question Answering Framework Based on RAG
Wu, R., Chen, S., Su, X., Zhu, Y., Liao, Y., & Wu, J. (2024) · International Conference on Information Science, Parallel and Distributed Systems (ISPDS) · doi:10.1109/ispds62779.2024.10667535 · Cited by 14
“…directly from the LLM) and SR-RAG (which retrieves relevant paragraphs based on the query), but also with several advanced RAG, including FL-RAG [13] , which employs previously set markers at intervals as query triggers for retrieval; FS-RAG [14] , which utilizes the last generated sentence as…”
MethodRetrieval-Pretrained Transformer: Long-range Language Modeling with Self-retrieval
Rubin, O., & Berant, J. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00693 · Cited by 44
Kamikubo, R., Kacorri, H., & Asakawa, C. (2024) · International Web for All Conference (W4A) · doi:10.1145/3677846.3677860 · Cited by 17
“Thus, to create more inclusive in-person recreational experiences for blind people, we aim next to delve into social navigation aspects [31, 63] within guidance systems.”
RAG-Enhanced Evidence Recommendation in Financial Legal Resolutions
Lee, H.-H., Chen, C.-C., & Yen, A.-Z. (2025) · International World Wide Web Conference (WWW) · doi:10.1145/3701716.3715520 · Cited by 8
Mask-based Membership Inference Attacks for Retrieval-Augmented Generation
Liu, M., Zhang, S., & Long, C. (2024) · arXiv · doi:10.48550/arxiv.2410.20142 · Cited by 8
ChatGPT, extended: large language models and the extended mind
Smart, P., Clowes, R., & Clark, A. (2025) · Synthese · doi:10.1007/s11229-025-05046-y · Cited by 13
“While Schwitzgebel et al. ( 2024 ) rely on fine-tuning to assimilate knowledge into the internal parameters of a GPT model, we rely on a technique, called Retrieval-Augmented Generation (RAG) (Gao et al., 2024 ; Lewis et al., 2020 ; Ram et al., 2023 ).”
Mask-based Membership Inference Attacks for Retrieval-Augmented Generation
Liu, M., Zhang, S., & Long, C. (2025) · International World Wide Web Conference (WWW) · doi:10.1145/3696410.3714771 · Cited by 54
Retrieval-augmented code completion for local projects using large language models
Hostnik, M., & Robnik‐Šikonja, M. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.128596 · Cited by 15
GINGER: Grounded Information Nugget-Based Generation of Responses
Łajewska, W., & Balog, K. (2025) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3726302.3730166 · Cited by 27
IRAGKR:Iterative retrieval augmented generation with fine-grained knowledge refinement
Du, K., Wang, W., Zhang, B., Wang, P., Zhang, F., Lin, C., & Guo, Y. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.131282 · Cited by 15
Grading Open‐Ended Questions Using LLMs and RAG
Rodriguez, J., Fernández‐García, A. J., & Verdú, E. (2025) · Expert Systems · doi:10.1111/exsy.70174 · Cited by 11
“Therefore, it could be investigated whether the REALM technique is more suitable to find a more relevant context, like this study (Ram et al. 2023) does.”
INKER: Adaptive dynamic retrieval augmented generation with internal-external knowledge integration
Zhou, M., Tang, J., Zeng, W., & Zhao, X. (2025) · Information Processing & Management · doi:10.1016/j.ipm.2025.104534 · Cited by 8
Using retrieval-augmented generation to elevate low-code developer skills
Наход, О. (2023) · Artificial Intelligence · doi:10.15407/jai2023.03.126 · Cited by 9
Predicting Missing Values in Survey Data Using Prompt Engineering for Addressing Item Non-Response
Ji, J., Kim, J., & Kim, Y. (2024) · Future Internet · doi:10.3390/fi16100351 · Cited by 12
Towards large language models with human-like episodic memory
Dong, C., Lu, Q., Norman, K. A., & Michelmann, S. (2025) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2025.06.016 · Cited by 28
Unlocking Multi-View Insights in Knowledge-Dense Retrieval-Augmented Generation
Chen, G., Yu, W., Lu, X., Zhang, X., Meng, E., & Sha, L. (2025) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2025.3622944 · Cited by 13
“These methods have shown effective improvements in the performance of LLM outputs by leveraging RAG’s ability to access and incorporate relevant information from external sources [16], [17].”
Leveraging RAG With Transformer for Context-Based Personalized Recommendations
Alamri, F. S., Rehman, A. U., Alghofaily, B., Ahmed, A. R., & Saleem, K. (2025) · IEEE Access · doi:10.1109/access.2025.3574073 · Cited by 13
“Ranking enhances information retrieval results, and the RAG pipeline has further improved generated content quality [17], [18].”
Yen, R., Xie, Y., Sultanum, N., & Zhao, J. (2025) · Designing Interactive Systems Conference · doi:10.1145/3715336.3735752 · Cited by 10
An Advanced Retrieval-Augmented Generation System for Manufacturing Quality ControlPreprint
Heredia, J. A., & Barreda, J. G. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4828863 · Cited by 6
L-Eval: Instituting Standardized Evaluation for Long Context Language Models
An, C., Gong, S., Zhong, M., Zhao, X., Li, M., Zhang, J., Kong, L., & Qiu, X. (2023) · arXiv · doi:10.48550/arxiv.2307.11088 · Cited by 6
Zhi, Z., Liu, Z., Elbadawi, M., Daneshmend, A., Orlu, M., Basit, A. W., Demosthenous, A., & Rodrigues, M. R. D. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.130502 · Cited by 24
Theuner, K., Elmgren, T. M., Götling, A., May, M. C., & Akay, H. (2025) · Procedia CIRP · doi:10.1016/j.procir.2025.03.073 · Cited by 7
Hermes: Algorithm-System Co-design for Efficient Retrieval-Augmented Generation At-Scale
Shen, M., Umar, M., Maeng, K., Suh, G. E., & Gupta, U. (2025) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1145/3695053.3731076 · Cited by 16
“In state-of-the-art RAG systems, multiple retrieval iterations occur per query to refresh the documents used for generating every 𝑠 tokens [39], a process known as retrieval striding.”
InfluentialGinting, M. F., Kim, D.-K., Kim, S. K., Krishna, B. J., Kochenderfer, M. J., Omidshafiei, S., & Agha–mohammadi, A. (2025) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra55743.2025.11128684 · Cited by 13
“Retrieval-augmented language model for robot planning : Retrieval-based approach for language model, such as augmented generation (RAG) [30], is a prominent framework for combining knowledge in a database with language generation [31], [32], [33].”
Efficient Tuning of Large Language Models for Knowledge-Grounded Dialogue Generation
Zhang, B., Ma, H., Li, D., Ding, J., Wang, J., Xu, B., & Lin, H. (2025) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl.a.17 · Cited by 7
Bailicai: A Domain-Optimized Retrieval-Augmented Generation Framework for Medical Applications
Cui, L., Liu, Y., Ouyang, C., Yu, Y., Zhang, G., Wan, Y., & Yang, F. (2026) · Big Data Mining and Analytics · doi:10.26599/bdma.2024.9020097 · Cited by 28
AISStream-MCP: A Real-Time Memory-Augmented Question-Answering System for Maritime Operations
Chen, S., Zhao, R., Yang, J., & Huang, Y. (2025) · Journal of Marine Science and Engineering · doi:10.3390/jmse13091754 · Cited by 7
Allies: Prompting Large Language Model with Beam Search
Sun, H., Liu, X., Gong, Y., Zhang, Y., Jiang, D., Yang, L., & Duan, N. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.247 · Cited by 20
“…task that usually requires access to external corpora [Petroni et al., 2021] and utilizes a retriever component for knowledge augmentation [Ram et al., 2023, Shi et al., 2023, Rashkin et al., 2021, Gao et al., 2022, Bohnet et al., 2022, Menick et al., 2022] to improve their ability to…”
BackgroundMethodkNN-LM Does Not Improve Open-ended Text Generation
Wang, S., Song, Y., Drozdov, A., Garimella, A., Manjunatha, V., & Iyyer, M. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.929 · Cited by 24
A RAG-based Question Answering System Proposal for Understanding Islam: MufassirQAS LLMPreprint
alan, A. Y., Karaarslan, E., & Aydın, Ö. (2024) · doi:10.36227/techrxiv.170723304.41988020/v2 · Cited by 9
Li, Y., Fu, X., Verma, G., Buitelaar, P., & Liu, M. (2025) · arXiv · doi:10.48550/arxiv.2510.24476 · Cited by 4
“Input-level integration, which concatenates retrieved documents with the query as input, remains the most widely adopted due to its simplicity and compatibility with existing models [115].”
R ag M e : Retrieval Augmented Video Generation for Enhanced Motion Realism
Peruzzo, E., Xu, D., Xu, X., Shi, H., & Sebe, N. (2025) · International Conference on Multimedia Retrieval · doi:10.1145/3731715.3733417 · Cited by 11
“Processing as a powerful way to improve model performances, by integrating information from an external database that acts as a memory bank [6, 29, 41].”
Wei, S., Hu, A., Wang, Z., Meng, X., Yu, L., Yang, B., & Qiu, J. (2025) · Journal of Applied Clinical Medical Physics · doi:10.1002/acm2.70263 · Cited by 5
An Agent-Based RAG Architecture for Intelligent Tourism Assistance: The Valencia Case Study
Bonetti, A., Salcedo-Puche, A., Vila‐Francés, J., García, X. B., Fernández-Vargas, E., Magdalena‐Benedito, R., & Soria‐Olivas, E. (2025) · Tourism and Hospitality · doi:10.3390/tourhosp6050266 · Cited by 5
“Retrieval has evolved from individual tokens (Khandelwal et al., 2020) and entities (B. Li et al., 2022) to more structured representations, including slices (Ram et al., 2023) and knowledge graphs (Kang et al., 2023), balancing granularity with precision and efficiency.”
EvaRAG: Evaluating Advanced RAG Techniques With Indexing and Distance Metrics
Elkiran, H., & Rasheed, J. (2025) · IEEE Access · doi:10.1109/access.2025.3646665 · Cited by 12
“RAG systems [12], [13] retrieve text to provide scope to LLMs. RAG pipelines use approximate nearest-neighbor (ANN) search methods to retrieve relevant documents from large databases.”
Navigating the Risks: A Survey of Security and Privacy Threats in LLM-Based Agents
Gan, Y., Yang, Y., W, Z., He, P., Zeng, R., Wang, Y., Li, Q., Zhou, C., Li, S., Wang, T., et al. (2026) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3807666 · Cited by 119
“Typical external knowledge can come from a retrieval database [26, 131, 168] or user input data [156, 174].”
Context Is King: Large Language Models’ Interpretability in Divergent Knowledge Scenarios
Piñeiro-Martín, A., Santos-Criado, F.-J., García-Mateo, C., Docío-Fernández, L., & López-Pérez, M. D. C. (2025) · Applied Sciences · doi:10.3390/app15031192 · Cited by 4
Large Language Models for SoC Security
Tehranipoor, M. M., Azar, K. Z., Asadizanjani, N., Rahman, F., Kamali, H. M., & Farahmandi, F. (2024) · Springer eBooks · doi:10.1007/978-3-031-58687-3_6 · Cited by 6
Effective Tool Augmented Multi-Agent Framework for Data Analysis
Zhang, X., Mao, Z., Chen, Z., & Gao, S. (2024) · Data Intelligence · doi:10.3724/2096-7004.di.2024.0013 · Cited by 17
“IRCoT [38] interleaves CoT generation with knowledge retrieval steps, while in-context RALM [39] enhances outputs by adding retrieved content to the input.”
BackgroundAutonomous Cyberattack with Security-Augmented Generative Artificial Intelligence
Gregory, J., & Liao, Q. (2024) · IEEE International Conference on Cyber Security and Resilience (CSR) · doi:10.1109/csr61664.2024.10679470 · Cited by 4
“Research [15] showed that in-context learning could be applied without modifying a LLM architecture, showing the possibilities of RAG as a form of drop-in model supplementation.”
BackgroundA Survey on Retrieval-Augmented Generation Technology in Large Language Models
Guo, J., Yang, H., & Liu, Q. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-4788-3_23 · Cited by 8
Standardizing Intelligence: Aligning Generative AI for Regulatory and Operational CompliancePreprint
Imperial, J. M., Jones, M. D., & Madabushi, H. T. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5146161 · Cited by 4
Farahani, M., & Johansson, R. (2024) · Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2024.emnlp-main.943 · Cited by 5
DRAGON: Enhancing On-Device Model Performance with Distributed Retrieval-Augmented Generation
Liu, S., Zheng, Z., Huang, X., Wu, F., Chen, G., & Wu, J. (2025) · ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc) · doi:10.1145/3704413.3764419 · Cited by 4
Structured reflective reasoning for precise medical knowledge graph retrieval augmented generation
Wang, B., Wu, J., Shi, Y., Chen, W., & Gong, F. (2025) · Health Information Science and Systems · doi:10.1007/s13755-025-00390-2 · Cited by 6
“Additionally, Ram et al. explored in-context retrieval-augmented language models, which dynamically incorporate retrieved information within the context window to enhance real-time generation [22].”
Fashion-RAG: Multimodal Fashion Image Editing via Retrieval-Augmented Generation
Sanguigni, F., Morelli, D., Cornia, M., & Cucchiara, R. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229186 · Cited by 5
“Additionally, emerging model-driven retrieval strategies [18], [34], [35] focus on dynamically optimizing both the selection and utilization of retrieved information.”
A zero-shot prompt learning approach on fine-grained text classification
Zhang, Z., Ge, L., Hu, R., & Wang, Y. (2026) · Scientific Reports · doi:10.1038/s41598-025-34825-3 · Cited by 5
“This pre-filtering step significantly reduces the cognitive load on the LLM and mitigates the risk of hallucinations caused by out-of-distribution data 34 .”
A context-augmented large language model for accurate precision oncology medicine recommendations
Jun, H., Tanaka, Y., Johri, S., Camp, S. Y., Bao, E. L., Carvalho, F. L. F., Gui, D. Y., Jordan, A. C., Labaki, C., Martin, S. D., et al. (2026) · Cancer Cell · doi:10.1016/j.ccell.2025.12.017 · Cited by 11
Chen, Z., Chen, X., Li, Z. B., Xue, J., & Gao, C. (2026) · Journal of Systems and Software · doi:10.1016/j.jss.2026.112855 · Cited by 6
Dong, L., Chen, Y., Ke, W., Wu, H., Deng, L., & Liao, G. (2026) · Information · doi:10.3390/info17040317 · Cited by 5
Park, S.-I., & Lee, J.-Y. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00724 · Cited by 23
“However, recent discoveries show that LLMs can be used as generators for RALM in an in-context setting, without additional training, by simply concatenating retrieved documents to the query (Levine et al., 2022b,a; Kamalloo et al., 2023; Shi et al., 2023b; Ram et al., 2023).”
Akbar, N. A. (2025) · BIO Web of Conferences · doi:10.1051/bioconf/202516701001 · Cited by 4
Santra, P., Ghosh, M., Ganguly, D., Basuchowdhuri, P., & Naskar, S. K. (2025) · Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery · doi:10.1002/widm.70021 · Cited by 7
From Domain Documents to Requirements: Retrieval-Augmented Generation in the Space Industry
Arora, C., Wang, F., Tantithamthavorn, C. K., et al. (2025) · IEEE International Requirements Engineering Conference (RE) · doi:10.1109/re63999.2025.00036 · Cited by 3
“3) An information retrieval model named ICRALM [9] will be applied to extended query Q|C Q to retrieval the related content from requirements documentation R .”
Wang, Y., Liu, Q., Fan, J., Hong, J., Chu, H., Tian, M., Gao, B., & Chen, H. (2026) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2026.3673470 · Cited by 17
“Furthermore, Ram et al. [41] study the pre-training and fine-tuning processes of RAG, demonstrating how RAG can further enhance model performance in the fine-tuning stage by incorporating large-scale external data sources during pre-training.”
Li, S., Wang, L., Sun, J., & Okatani, T. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-78183-4_29 · Cited by 10
How can language models assist with pharmaceuticals manufacturing deviations and investigations?
Salami, H., Smith, B. M., & Yadav, V. (2024) · International Journal of Pharmaceutics · doi:10.1016/j.ijpharm.2024.125100 · Cited by 7
GEM-RAG: Graphical Eigen Memories for Retrieval Augmented Generation
Rappazzo, B., Wang, Y., Ferber, A., & Gomes, C. P. (2024) · International Conference on Machine Learning and Applications · doi:10.1109/icmla61862.2024.00196 · Cited by 5
“Motivated by this problem, retrieval augmented generation (RAG) has been an increasingly active area of research, which aims to adapt static LLMs to new and niche domain applications [7], [8], [9], [10], [11].”
Empower Vision Applications with LoRA LMM
Mi, L., Wang, W., Tu, W., He, Q., Kong, R., Fang, X., Dong, Y., Zhang, Y., Li, Y., Li, M., et al. (2025) · European Conference on Computer Systems · doi:10.1145/3689031.3717472 · Cited by 50
Data Imputation Based on Retrieval-Augmented Generation
Shi, X., Wang, J., Chung, G., Julian, D., & Qiao, L. (2025) · Applied Sciences · doi:10.3390/app15137371 · Cited by 4
CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion
Yao, J., Li, H., Liu, Y., Ray, S., Cheng, Y., Zhang, Q., Du, K., Lu, S., & Jiang, J. (2026) · ACM Transactions on Computer Systems · doi:10.1145/3790254 · Cited by 3
“Retrieval augmented generation (RAG): RAG [22, 23, 37, 46, 48] can enhance the accuracy and reliability of LLMs with text chunks fetched from external sources.”
Metadata
- Authors
- Ram, O., Levine, Y., Dalmedigos, I., Muhlgay, D., Shashua, A., Leyton-Brown, K., & Shoham, Y.
- Affiliations
- AI21 Labs, Israel. orir@ai21.com
- AI21 Labs, Israel. yoavl@ai21.com
- AI21 Labs, Israel. itayd@ai21.com
- AI21 Labs, Israel. dorm@ai21.com
- AI21 Labs, Israel. amnons@ai21.com
- AI21 Labs, Israel. kevinlb@ai21.com
- AI21 Labs, Israel. yoavs@ai21.com
- Corresponding author
- Ram, Ori
- Levine, Yoav
- Dalmedigos, Itay
- Muhlgay, Dor
- Shashua, Amnon
- Shoham, Yoav
- Venue
- Transactions of the Association for Computational Linguistics
- Publisher
- MIT Press
- Published
- 2023
- Volume
- 11
- Pages
- 1316-1331
- ISSN
- 2307-387X (ISSN-L)
- OpenAlex
- W4388778348
- Cited by
- 449
- References
- 55
- Citations by year
- 2026: 160 · 2025: 189 · 2024: 80 · 2023: 20
- FWCI
- 42.21
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Transactions of the Association for Computational Linguistics (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion, public-domain)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University) (public-domain)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- retrieval-augmented language models, document retrieval, document ranking, factual accuracy, source attribution, pre-trained language models, large language models
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.