Latest Retrieval-Augmented Generation Research Papers

The newest Retrieval-Augmented Generation papers from across the field — arXiv, NeurIPS, CVPR, Nature, and more — refreshed daily and ranked by relevance. Distill AI tracks Retrieval-Augmented Generation so you don’t have to: get the standout work delivered to your inbox every morning, with 2-sentence summaries and the option to chat with any paper.

Recent papers

• AI Systems for Qualitative Data Interrogation (EpiChat) and Semi-Automated Coding (EpiCode)
  Tepy Sokun Chriv, Panha Lim, Cheaminh Chiep, Sophea Iv et al. · Zenodo (CERN European Organ... · Nov 3, 2026

  Background: The Virage project collected approximately 800 one-hour interviews on rabies awareness, including transmission, disease progression, and vaccination. Manually searching, comparing, and coding this corpus would require substantia…

• Homebrewed SLMs: Autoregressive and Masked Diffusion Language Models Built at Home for Portuguese
  Pedro Carvalho Brom · arXiv · Oct 29, 2026

  Building a language model of one's own no longer requires a data centre. This tutorial trains one from scratch, tokenizer included, on one consumer GPU, and shows how to tell whether it is good for the tasks of a specialist domain. It propo…

• FIRME: Registry-Anchored Memory Graphs for Evolving, Heterogeneous Investor Knowledge
  Yunbai Zhang, Yunyi Gao, Han-Tai Shiao · arXiv · Oct 29, 2026

  Financial AI workflows that reuse investor knowledge need persistent memory that integrates observations across sources and periods without losing identity or provenance. In a knowledge graph, recurring observations may use different surfac…

• Glance: The Generator Is Its Own Reranker
  Adilet Metinov, Chingiz Arziev · arXiv · Oct 29, 2026

  Retrieval-augmented generation commonly serves separate models for reranking and answering, duplicating model residency and passage processing. We introduce Glance, which equips the generator with a lightweight decision adapter and pointer …

• ScienceJudge: Pattern Perturbation Augmentation for Sentence-Level Scientific Claim Verification
  Chuhai Cai, Xin Chen, Min Yang, Feng Jiang · arXiv · Oct 29, 2026

  Sentence-level scientific claim verification determines whether each sentence in an AI-generated scientific paragraph is supported by a given multimodal evidence bundle. Fine-grained reporting errors often preserve fluent and plausible stat…

• Towards Full-Text Scholarly Argument Mining using Entity-Anchored Claim Extraction
  Bhuvanesh Verma, Andy Lüking, Mounika Marreddy, Leon Lukas Hammerla et al. · arXiv · Oct 29, 2026

  Argument mining in full-text scholarly articles remains an important yet challenging task. This difficulty stems not only from the need to model long-distance dependencies but also from the substantial effort required to annotate sentence-l…

• Hallucination Risk in Retrieval-Augmented Generation: A Probabilistic Model of Retrieval Recall, Context Length and Answer Faithfulness
  Wanjiru Njeri Kamau, Brian Otieno Odhiambo · Zenodo (CERN European Organ... · Oct 28, 2026
• Artificial intelligence applications in forensic science and their implications for education and training
  María Belén Morales-Cevallos, Canva Byron Ma Lam, Kevin David Dickens Guerrero, Diego Fabián Vique López · Frontiers in Education · Oct 28, 2026

  Introduction Artificial intelligence (AI) is becoming increasingly relevant to forensic practice and has potential applications in forensic education, including evidence interpretation, simulation-based learning, and decision-making. Then, …

• ResearchMind AI An AI-Powered Multi-Domain Research Paper Analysis and Report Generation System
  Mr Sanjay Krishna B, Anandhi A, Karthikha shree S M, Lekha P et al. · Zenodo (CERN European Organ... · Oct 28, 2026

  ResearchMind AI is an AI-powered system designed to simplify research paper analysis across multiple domains. It enables users to upload and analyze multiple research papers in PDF format. PyMuPDF is used to extract text, while Sentence Tra…

• UCVE: Graph Structured Retrieval for Compliance Validation
  Emily Chen, Michelle Menachery, Himanshu Kumar, Rahul Nenavath et al. · arXiv · Oct 27, 2026

  Structured compliance validation requires determining whether every obligation in a predefined requirement space is satisfied, violated, or unaddressed by a given document. This can be described as a coverage problem that existing retrieval…

• Coverage Under Contradiction: A Benchmark for Negation-Aware Retrieval
  Pasha Abrishamchian, Eugene Yang, Andrew Yates, Nazli Goharian et al. · arXiv · Oct 27, 2026

  Coverage-evaluation benchmarks quantify whether retrieved evidence covers the different pieces of information needed to satisfy an information need. Existing benchmarks mostly consider factual claims that may already be known to pretrained …

• SpectRAG: Weighted Chunk Graphs for Token-Budgeted Context Selection in Multi-Hop Retrieval-Augmented Generation
  Zafer Sahinoglu, Abdulazeez Mohammed, Buse Yaren Kazangirler, Cahit Berkay Kazangirler · arXiv · Oct 27, 2026

  Retrieval-augmented generation (RAG) systems ground large language models (LLMs) by retrieving external evidence and conditioning generation on the retrieved context. Although dense retrieval is effective for local semantic matching, indepe…

• The State and Fate of Multilingual, Contextual Evaluation in the NLP World
  Manan Uppadhyay, Himanshu Beniwal, Prashant Kodali, Sunayana Sitaram · COLM · Oct 27, 2026

  Multilingual evaluation benchmarks are the primary instrument for assessing whether large language models generalize beyond English, yet the adequacy of these benchmarks has received little systematic scrutiny. We present a data-driven audi…

• Self-Keyed In-Parameter Memorization for External Memory-Free Dialogue Recall in Frozen Language Models
  XinYu Piao, Myunghoon Ryu, Joongheon Kim, Jong-Kook Kim · arXiv · Oct 27, 2026

  A large language model maintains conversational content in an external storage that is outside the model and grows as the dialogue continues. This work assumes an environment where a language model has no external memory and no prior knowle…

• TrawMem: Reframing Agent Memory as Task-Oriented Role-Adaptive Workspaces
  Mingyu Zhang, Zixu Li, Zhiheng Fu, Zhiwei Chen et al. · arXiv · Oct 27, 2026

  Large language model (LLM) agents rely on memory systems to leverage long-term interactions, yet existing systems typically retrieve memories based on their relevance to the current query and use them as contextual evidence. However, releva…

• A Framework for Finished Goods Safety Stock Estimation Using Local LLM-Based Retrieval-Augmented Generation (RAG)
  Hyoungwoo Cho, Woo‐Sik Yoo · Journal of Korean Institute... · Oct 27, 2026

  This study proposes a local LLM-RAG-based framework for estimating finished-goods safety stock in high-mix low-volume manufacturing environments. The proposed framework maintains a statistical safety stock calculation as a quantitative anch…

• Contextual Compression with Attention: Enhancing Input Efficiency for Long-Contexts in Retrieval-Augmented Generation
  Lee JunSeo, Yewon Lee, Dooyoung Kim, Youngjoong Ko · Journal of KIISE · Oct 27, 2026

  검색증강생성(RAG)은 질의와 관련된 외부 지식을 언어 모델의 입력으로 사용하여 거대 언어 모델(LLM)의 환각 문제를 해결하고자 고안된 방법이다. 하지만 LLM은 추론 속도가 느리고 입력된 문서의 길이에 따라 모델의 연산량이 증가하므로 대규모 문서를 처리하는데 한계가 있다. 본 연구는 이러한 문제점을 극복하기 위해, LLM의 입력 효율성을 개선하는 ‘어텐션 기반 문맥 압축(Contextual Compression with A…

• Fidelidad de cita en RAG empresarial: por qué recuperar un pasaje no es apoyarlo y por qué la cita adjunta aumenta la confianza del lector antes que la corrección de la respuesta (2023–2026) / Citation fidelity in enterprise RAG: why retrieving a passage is not supporting it and why the attached citation raises reader trust before it raises answer correctness (2023–2026)
  Santiago Pineda Botero · Zenodo (CERN European Organ... · Oct 27, 2026

  La generación aumentada por recuperación se instaló entre 2023 y 2026 como arquitectura por defecto del asistente interno de conocimiento sobre la premisa de que inyectar el documento suprime la invención. Este working paper sostiene que la…

• Offline Language-Based Support System for Tactical Decision-Making in Air Combat Operations
  Joao P. A. Dantas · arXiv · Oct 26, 2026

  An offline language-based support tool is presented to assist pilot training and tactical decision-making in air combat operations. The system, named MATE (Mission Assistant for Training and Evaluation), applies retrieval-augmented generati…

• CalibraRL: Confidence-Calibrated Reinforcement Learning for Mitigating Hallucinations in RAG
  Qianchi Zhang, Hainan Zhang, Liang Pang, Hongwei Zheng et al. · arXiv · Oct 26, 2026

  Large language models (LLMs) often produce plausible but incorrect answers, motivating reinforcement learning methods that reward correct answers, penalize incorrect answers, and allow abstention. However, fixed outcome-based rewards do not…

• RAGing Against Misogyny: Improving Implicit Misogyny Explanations
  Ivana Crescenzi, Arianna Muti, Maarten Sap, Tommaso Caselli et al. · arXiv · Oct 26, 2026

  Implicit misogyny often requires background knowledge to be understood. To address this challenge, we propose a Retrieval-Augmented Generation (RAG) framework for detecting and explaining implicit misogyny in English and Italian, leveraging…

• RAG-Based MCA AI Agent for Document-Grounded College Information Assistance Using Retrieval-Augmented Generation, Vector Search and a Locally Hosted Large Language Model
  Nillesh Patil, Sainath Bejgamwar, Keyur Ingale, Manas Gadkari et al. · Zenodo (CERN European Organ... · Oct 25, 2026

  This paper presents the design and implementation of a Retrieval-Augmented Generation (RAG)-based MCA AI Agent for document-grounded question answering over college and student policy documents. The system uses PyMuPDF for PDF text extracti…

• GraphRAG Retrieval of Regulatory Knowledge: An Empirical Evaluation for Ecuadorian Tax Legislation Query Answering
  Daniel Vinicio Cuatín Flores, Jorge Andrés Mosquera Sotomayor, Carlos Iván Díaz Hidalgo, Pablo Marcel Recalde Varela et al. · Applied Sciences · Oct 25, 2026

  Tax regulation in Ecuador changes constantly, and the resulting information is scattered across laws, regulations, resolutions, forms, and annexes with strong ties across meaning and time. We compare four retrieval-augmented generation (RAG…

• Cumulative Memory Sizing and Asymmetric Latency Dynamics in Local Retrieval-Augmented Generation Pipelines (2026)
  Abdulrahman Maslmany · Zenodo (CERN European Organ... · Oct 25, 2026

  This systems engineering paper models end-to-end memory allocation and pipeline execution latency in air-gapped, on-premises Retrieval-Augmented Generation (RAG) architectures. A persistent operational bottleneck in local AI deployment is i…

• A Layout-Aware Hybrid Retrieval-Augmented Generation Framework for Mitigating Hallucination in Turkish Texts: Evidence from the Turkish Official Gazette
  Samet Sarıyıldız, Mustafa Yağcı · Applied Sciences · Oct 25, 2026

  Large language models (LLMs) hallucinate, which limits their use in domains with little tolerance for error, such as law. We present a layout-aware hybrid retrieval-augmented generation (RAG) system for answering questions over 30 issues of…

• LLM-BASED SEMANTIC RESUME RANKING AND CANDIDATE RECOMMENDATION USING RETRIEVAL-AUGMENTED GENERATION
  International Research Jour... · Oct 25, 2026

  Shortlisting candidates for a vacancy means reading many profiles against one job description and deciding who is worth a recruiter's time.Keyword filters miss candidates who describe the same skill in different words, and a language model …

• RAG-ассистент по базе знаний компании: как сотрудники находят ответы за секунды
  Alexander Sverdlov · Zenodo (CERN European Organ... · Oct 25, 2026

  Третий день работы нового менеджера. Клиент спрашивает, можно ли вернуть товар без чека. Ответ где-то записан: в регламенте на общем диске, в закреплённом сообщении рабочего чата или в письме руководителя двухлетней давности. Менеджер делае…

• StudyGenAI: A Comprehensive Review of Multimodal Generative AI and Spaced Repetition in Modern EdTech
  Saujanya Sachin Ganbavale, Priti Gangaram Gavade, Shravani Satish Chougule, Shrutika Sambhaji Kumbhar · Zenodo (CERN European Organ... · Oct 25, 2026

  Traditional education heavily relies on passive reading from static textbooks and unstructured handwritten notes, leading to inefficient study habits and poor long-term retention. This review explores the paradigm shift from standard Learni…

• A MULTI-FORMAT RETRIEVAL-AUGMENTED GENERATION AGENT FOR CONVERSATIONAL DOCUMENT QUESTION ANSWERING
  Umair Aslam,Wajid Ali,Muhammad Ahmad Hanif,Hifza Rani,Basit Bashir,Muhammad Yousif · Zenodo (CERN European Organ... · Oct 24, 2026
• Provenance-audited retrieval-augmented generation for national adaptation reporting: a computer-science evaluation of citation-fabrication risk in climate analytics pipelines
  Amina Aït Benhaddou, Mohamed Boussaidane Hicham Benomar · Zenodo (CERN European Organ... · Oct 24, 2026

  National adaptation reporting under the Paris Agreement requires climate-analytics pipelines that can trace every quantitative claim to a verifiable source. Retrieval-augmented generation (RAG) offers a way to ground narrative synthesis in …