RAG / 检索增强生成

End-to-End Training of Multi-Document Reader and Retriever for Open-Domain Question Answering

Devendra Singh Sachan, Siva Reddy, William Hamilton, Chris Dyer, Dani Yogatama

专题命中 知识库问答 ：retriever(title,abstract)；分类 cs.IR、cs.CL、cs.AI

Comments NeurIPS 2021 camera-ready version

RAG-based Question Answering over Heterogeneous Data and Text

Philipp Christmann, Gerhard Weikum

专题命中 知识库问答 ：RAG(title,abstract)；分类 cs.IR、cs.CL

Comments IEEE Data Engineering Bulletin -- December 2024 Edition on RAG

EviBack: Search-Agent Reinforcement Learning via Evidence-Constrained Teacher Backoff

EviBack：通过证据约束的教师退避进行搜索智能体强化学习

Xiao Ma, Zhiquan Hu, Yi Wei, Chenchen Zhao, Yijun Chen, Jicheng Zhao, Yuming Li, Chuang Dai

专题命中 知识库问答 ：RAG(summary_cn,abstract)；分类 cs.AI

AI总结 研究针对智能体RAG系统中全零展开组问题，提出EviBack方法，通过证据约束教师退避提供辅助监督。利用全自动管道生成两阶段教师，提升了下游F1等指标，在多个问答基准上取得更好效果。

RASER: Recoverability-Aware Selective Escalation Router for Multi-Hop Question Answering

RASER: 可恢复性感知的选择性升级路由器用于多跳问答

Yuyang Li, Zihe Yan, Tobias Käfer

机构 * Institute AIFB, Karlsruhe Institute of Technology, Karlsruhe, Germany（卡尔斯鲁厄理工学院AIFB研究所） ； Shanghai Jiao Tong University, Shanghai, China（上海交通大学） 纠错

专题命中 知识库问答 ：RAG(summary_cn,abstract)；分类 cs.AI

AI总结 提出RASER路由器，基于单次RAG的六个特征决定是否升级到更昂贵的检索策略，在不增加额外LLM调用的情况下，在F1分数与SOTA相当的同时节省大量token。

Comments Under Review

Bridging Dual Knowledge Graphs for Multi-Hop Question Answering in Construction Safety

连接双知识图谱以实现施工安全多跳问答

Yuxin Zhang, Xi Wang, Mo Hu, Zhenyu Zhang

机构 * organization= Department of Construction Science, College of Architecture, Texas A\&M University, College Station , country= USA 纠错

专题命中 知识库问答 ：RAG(abstract,abstract_cn)；retrieval-augmented generation(abstract)；hybrid retrieval(abstract)；分类 cs.AI

AI总结 本文提出BifrostRAG系统，通过双图检索增强生成模型处理施工安全多跳问答，实现92.8%精度和85.5%召回率，优于传统基线方法。

Comments 22 pages, 13 figures

Journal ref Automation in Construction, Volume 183, March 2026, 106794

RPDR: A Round-trip Prediction-Based Data Augmentation Framework for Long-Tail Question Answering

RPDR：基于回程预测的数据增强框架用于长尾问答

Yiming Zhang, Siyue Zhang, Junbo Zhao, Chen Zhao

机构 * Zhejiang University（浙江大学） ； Nanyang Technological University（南洋理工大学） ； NYU Shanghai（纽约大学上海分校） ； Center for Data Science, New York University（纽约大学数据科学中心） 纠错

专题命中 知识库问答 ：retrieval-augmented generation(abstract)；RAG(abstract)；retriever(abstract)；dense retrieval(abstract)

AI总结 RPDR通过数据增强框架提升长尾问答性能，利用回程预测选择易学实例并动态路由查询以优化检索效果。

Decomposing Retrieval Failures in RAG for Long-Document Financial Question Answering

在长文档金融问答中分解检索失败

Amine Kobeissi, Philippe Langlais

专题命中 知识库问答 ：RAG(title)；retrieval-augmented generation(abstract)；分类 cs.IR、cs.CL

AI总结 本文针对长文档金融问答中的检索失败问题，提出一种基于页面的分层检索方法，通过微调双编码器提升页面和片段的检索效果。

From Static to Dynamic: A Streaming RAG Approach to Real-time Knowledge Base

Yuzhou Zhu

专题命中 知识库问答 ：RAG(title,abstract)；分类 cs.IR、cs.AI

ComRAG: Retrieval-Augmented Generation with Dynamic Vector Stores for Real-time Community Question Answering in Industry

Qinwen Chen, Wenbiao Tao, Zhiwei Zhu, Mingfan Xi, Liangzhong Guo, Yuan Wang, Wei Wang, Yunshi Lan

机构 * School of Data Science and Engineering, East China Normal University（数据科学与工程学院，东华大学） ； Alibaba Group（阿里巴巴集团） 纠错

专题命中 知识库问答 ：retrieval-augmented generation(title,abstract)；分类 cs.CL、cs.AI

Comments 7 pages, 4 figures. Accepted at ACL 2025 Industry Track

Evidence-backed Fact Checking using RAG and Few-Shot In-Context Learning with LLMs

Ronit Singhal, Pransh Patwa, Parth Patwa, Aman Chadha, Amitava Das

专题命中 知识库问答 ：RAG(title,abstract)；分类 cs.CL、cs.AI

Comments Accepted in The Seventh FEVER Workshop at EMNLP 2024

A Multimodal Dense Retrieval Approach for Speech-Based Open-Domain Question Answering

Georgios Sidiropoulos, Evangelos Kanoulas

专题命中 知识库问答 ：dense retrieval(title)；retriever(abstract)；分类 cs.IR、cs.CL

Towards Universal Dense Retrieval for Open-domain Question Answering

Christopher Sciavolino

专题命中 知识库问答 ：dense retrieval(title,abstract)；分类 cs.IR、cs.CL

GARAGE: Characterizing the Automation Boundary in LLM-based Attack Graph Generation

GARAGE：基于大语言模型的攻击图生成中自动化边界的特征描述

Daekwon Pi, Sangho Lee, Young Hun Lee, Huy Kang Kim

专题命中 知识库问答 ：RAG(summary_cn,abstract)

AI总结 研究针对现代车辆安全CTI合成难题，提出GARAGE框架，通过RAG技术将碎片化CTI转化为特定领域知识库用于攻击图生成，经实验验证能准确转移安全知识，还可作为TARA支持工具提供性价比分析以指导在各LLM层级部署。

Comments 22 pages, 10 figures, 12 tables

Learning from Natural Language Feedback for Personalized Question Answering

通过自然语言反馈学习实现个性化问答

Alireza Salemi, Hamed Zamani

机构 * Center for Intelligent Information Retrieval（智能信息检索中心） ； University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校） 纠错

专题命中 知识库问答 ：RAG(abstract,abstract_cn)；retrieval-augmented generation(abstract)；分类 cs.IR、cs.CL、cs.AI

AI总结 本文提出VAC框架，利用自然语言反馈替代标量奖励，提升个性化问答效果，实验表明其在LaMP-QA基准上表现优异。

Efficient Context Selection for Long-Context QA: No Tuning, No Iteration, Just Adaptive-$k$

Chihiro Taguchi, Seiji Maekawa, Nikita Bhutani

机构 * University of Notre Dame（内布拉斯加大学达灵顿分校） ； Megagon Labs（梅加贡实验室） 纠错

专题命中 知识库问答 ：retrieval-augmented generation(abstract)；RAG(abstract)；retriever(abstract)；分类 cs.IR、cs.CL、cs.AI

Comments 26 pages, 16 tables, 5 figures. Accepted at EMNLP 2025 (Main)

Reinforcement Learning Integrated Agentic RAG for Software Test Cases Authoring

强化学习集成的代理RAG用于软件测试用例编写

Mohanakrishnan Hariharan

专题命中 知识库问答 ：RAG(title,abstract)；分类 cs.AI

AI总结 本文提出一种结合强化学习与自主代理的RAG框架，用于提升软件测试用例生成的准确性和缺陷检测率。

Chronological Passage Assembling in RAG framework for Temporal Question Answering

Byeongjeong Kim, Jeonghyun Park, Joonho Yang, Hwanhee Lee

机构 * Department of Artificial Intelligence, Chung-Ang University（人工智能系， Chung-Ang 大学） 纠错

专题命中 知识库问答 ：RAG(title,abstract)；分类 cs.CL

Comments 15 pages, 4 figures

BYOKG-RAG: Multi-Strategy Graph Retrieval for Knowledge Graph Question Answering

Costas Mavromatis, Soji Adeshina, Vassilis N. Ioannidis, Zhen Han, Qi Zhu, Ian Robinson, Bryan Thompson, Huzefa Rangwala, George Karypis

机构 * Amazon（亚马逊） 纠错

专题命中 知识库问答 ：RAG(title,abstract)；分类 cs.CL

From Unstructured Communication to Intelligent RAG: Multi-Agent Automation for Supply Chain Knowledge Bases

Yao Zhang, Zaixi Shang, Silpan Patel, Mikel Zuniga

机构 * Amazon Operational Technology Solutions（亚马逊运营技术解决方案） 纠错

专题命中 知识库问答 ：RAG(title,abstract)；分类 cs.AI

Comments Accepted In Proceedings of the 1st Workshop on AI for Supply Chain: Today and Future @ 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2 (KDD 25), August 3, 2025, Toronto, ON, Canada. ACM, New York, NY, USA, 14 pages, 2 figures

E^2GraphRAG: Streamlining Graph-based RAG for High Efficiency and Effectiveness

Yibo Zhao, Jiapeng Zhu, Ye Guo, Kangkang He, Xiang Li

机构 * School of Data Science and Engineering, East China Normal University（数据科学与工程学院，东华大学） ； China Baowu Group（宝武集团） 纠错

专题命中 知识库问答 ：RAG(title,abstract)；分类 cs.AI

Comments 16 pages

LoRE: Logit-Ranked Retriever Ensemble for Enhancing Open-Domain Question Answering

Saikrishna Sanniboina, Shiv Trivedi, Sreenidhi Vijayaraghavan

专题命中 知识库问答 ：retriever(title,abstract)；分类 cs.CL

RetrievalQA: Assessing Adaptive Retrieval-Augmented Generation for Short-form Open-Domain Question Answering

Zihan Zhang, Meng Fang, Ling Chen

专题命中 知识库问答 ：retrieval-augmented generation(title,abstract)；分类 cs.CL

Comments Findings of ACL 2024

Generator-Retriever-Generator Approach for Open-Domain Question Answering

Abdelrahman Abdallah, Adam Jatowt

专题命中 知识库问答 ：retriever(title,abstract)；分类 cs.CL

Mitigating the Impact of False Negatives in Dense Retrieval with Contrastive Confidence Regularization

Shiqi Wang, Yeqin Zhang, Cam-Tu Nguyen

专题命中 知识库问答 ：dense retrieval(title,abstract)；分类 cs.CL

Comments Accepted by AAAI24

Task-Aware Specialization for Efficient and Robust Dense Retrieval for Open-Domain Question Answering

Hao Cheng, Hao Fang, Xiaodong Liu, Jianfeng Gao

专题命中 知识库问答 ：dense retrieval(title,abstract)；分类 cs.CL

Low-Resource Dense Retrieval for Open-Domain Question Answering: A Comprehensive Survey

Xiaoyu Shen, Svitlana Vakulenko, Marco del Tredici, Gianni Barlacchi, Bill Byrne, Adrià de Gispert

专题命中 知识库问答 ：dense retrieval(title,abstract)；分类 cs.CL

Combining Lexical and Dense Retrieval for Computationally Efficient Multi-hop Question Answering

Georgios Sidiropoulos, Nikos Voskarides, Svitlana Vakulenko, Evangelos Kanoulas

专题命中 知识库问答 ：dense retrieval(title,abstract)；分类 cs.IR

Comments Accepted at the 2nd Workshop on Simple and Efficient Natural Language Processing (SustaiNLP 2021)

Multi-step Retriever-Reader Interaction for Scalable Open-domain Question Answering

Rajarshi Das, Shehzaad Dhuliawala, Manzil Zaheer, Andrew McCallum

专题命中 知识库问答 ：retriever(title,abstract)；分类 cs.CL

Comments Published at ICLR 2019

Learning to Attend On Essential Terms: An Enhanced Retriever-Reader Model for Open-domain Question Answering

Jianmo Ni, Chenguang Zhu, Weizhu Chen, Julian McAuley

专题命中 知识库问答 ：retriever(title,abstract)；分类 cs.CL

LakeQuest: A Three-Domain Benchmark for Grounded Question Answering across Data Lakes

LakeQuest：用于跨数据湖的有基础问答的三领域基准测试

Michael Solodko, Steven Gong, Guangwei Yu, Satya Krishna Gorti, Jesse C. Cresswell, Victor Zhong

机构 * University of Waterloo（滑铁卢大学） 纠错

专题命中 知识库问答 ：RAG(abstract,abstract_cn)；retrieval-augmented generation(abstract)；分类 cs.CL、cs.AI

AI总结 介绍LakeQuest基准测试，用于评估跨数据湖的有基础问答。它跨越三个领域，含9846个QA对及证据指针，能暴露系统故障模式。通过基线评估发现高质量检索不能保证正确推理，凸显未来智能QA系统需强大发现和跨文件组合机制。

Comments 24 pages, 4 figures, 18 tables. Accepted at the Conference on Language Modeling (COLM) 2026