Applications Taxonomy 应用场景分类

RAG for Text 文本领域检索增强生成

• Question Answering 问答任务

  Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering 结合段落检索与生成模型实现开放域问答

  REALM: Retrieval-Augmented Language Model Pre-Training REALM：检索增强型语言模型预训练

  Knowledge Graph Based Synthetic Corpus Generation for Knowledge-Enhanced Language Model Pre-training 面向知识增强语言模型预训练的知识图谱合成语料构建

  Atlas: Few-shot Learning with Retrieval Augmented Language Models Atlas：基于检索增强语言模型的小样本学习

  Improving Language Models by Retrieving from Trillions of Tokens 依托万亿级词元检索优化语言模型

  Self-Knowledge Guided Retrieval Augmentation for Large Language Models 基于自知识引导的大语言模型检索增强

  Knowledge-Augmented Language Model Prompting for Zero-Shot Knowledge Graph Question Answering 面向零样本知识图谱问答的知识增强提示方法

  Think-on-Graph: Deep and Responsible Reasoning of Large Language Model with Knowledge Graph 图上推理：结合知识图谱实现大模型深度可信推理

  Nonparametric Masked Language Modeling 非参数掩码语言建模

  CL-ReLKT: Cross-lingual Language Knowledge Transfer for Multilingual Retrieval Question Answering CL-ReLKT：面向多语言检索问答的跨语言知识迁移

  One Question Answering Model for Many Languages with Cross-lingual Dense Passage Retrieval 融合跨语言稠密段落检索的多语言统一问答模型

  Entities as Experts: Sparse Memory Access with Entity Supervision 实体即专家：基于实体监督的稀疏内存访问机制

  When to Read Documents or QA History: On Unified and Selective Open-domain QA 文档与问答历史动态选用：统一选择性开放域问答研究

  Enhancing LLM Intelligence with ARM-RAG: Auxiliary Rationale Memory for Retrieval Augmented Generation ARM-RAG：借助辅助推理记忆提升检索增强生成效果

  DISC-LawLLM: Fine-tuning Large Language Models for Intelligent Legal Service DISC-LawLLM：面向智能法务的大语言模型微调

• Fact verification 事实核验

  CONCRETE: Improving Cross-lingual Fact-checking with Cross-lingual Retrieval CONCRETE：利用跨语言检索优化跨语言事实核验

  Stochastic RAG: End-to-End Retrieval-Augmented Generation through Expected Utility Maximization 随机检索增强生成：基于期望效用最大化的端到端RAG方案

• Commonsense Reasoning 常识推理

  KG-BART: Knowledge Graph-Augmented {BART} for Generative Commonsense Reasoning KG-BART：知识图谱增强BART实现生成式常识推理

  What Evidence Do Language Models Find Convincing? 语言模型认可的有效证据探究

  Enhancing Financial Sentiment Analysis via Retrieval Augmented Large Language Models 基于检索增强大模型优化金融情感分析

• Human-Machine Conversation 人机对话

  Grounded Conversation Generation as Guided Traverses in Commonsense Knowledge Graphs 依托常识知识图谱路径引导实现知识锚定对话生成

  Skeleton-to-Response: Dialogue Generation Guided by Retrieval Memory 骨架生成回复：检索记忆引导的对话生成

  Internet-Augmented Dialogue Generation 互联网知识增强对话生成

  BlenderBot 3: a deployed conversational agent that continually learns to responsibly engage BlenderBot 3：可持续自主学习的落地式合规对话智能体

  A Model of Cross-Lingual Knowledge-Grounded Response Generation for Open-Domain Dialogue Systems 面向开放域对话系统的跨语言知识锚定回复生成模型

  From Classification to Generation: Insights into Crosslingual Retrieval Augmented ICL 从分类到生成：跨语言检索增强上下文学习研究启示

  Cross-Lingual Retrieval Augmented Prompt for Low-Resource Languages 面向小语种的跨语言检索增强提示

  Citation-Enhanced Generation for LLM-based Chatbot 基于大模型聊天机器人的引用增强生成

  KAUCUS: Knowledge Augmented User Simulators for Training Language Model Assistants KAUCUS：用于训练大模型助手的知识增强用户模拟框架

• Neural Machine Translation 神经机器翻译

  Neural Machine Translation with Monolingual Translation Memory 融合单语翻译记忆的神经机器翻译

  Nearest Neighbor Machine Translation 近邻检索机器翻译

  Training Language Models with Memory Augmentation 结合记忆增强的语言模型训练

• Event Extraction 事件抽取

  Retrieval-Augmented Generative Question Answering for Event Argument Extraction 面向事件论元抽取的检索增强生成式问答

• Summarization 文本摘要

  Retrieval-Augmented Multilingual Keyphrase Generation with Retriever-Generator Iterative Training 检索器-生成器迭代训练的检索增强多语言关键词生成

  Unlimiformer: Long-Range Transformers with Unlimited Length Input Unlimiformer：支持超长输入的长距离Transformer模型

  Retrieval-based Full-length Wikipedia Generation for Emergent Events 面向热点事件的检索式维基百科全文生成

  RIGHT: Retrieval-augmented Generation for Mainstream Hashtag Recommendation RIGHT：面向热门标签推荐的检索增强生成

  M-RAG: Reinforcing Large Language Model Performance through Retrieval-Augmented Generation with Multiple Partitions M-RAG：基于多分区检索增强生成提升大模型性能

RAG for Code 代码领域检索增强生成

• Code Generation 代码生成

  Retrieval-Based Neural Code Generation 基于检索的神经代码生成

  Retrieval Augmented Code Generation and Summarization 检索增强代码生成与代码摘要

  When Language Model Meets Private Library 语言模型与私有代码库的融合方案

  Language Models of Code are Few-Shot Commonsense Learners 代码语言模型具备小样本常识学习能力

  DocPrompting: Generating Code by Retrieving the Docs 文档提示：依托文档检索实现代码生成

  CodeT5+: Open Code Large Language Models for Code Understanding and Generation CodeT5+：面向代码理解与生成的开源代码大模型

  AceCoder: Utilizing Existing Code to Enhance Code Generation AceCoder：利用已有代码提升代码生成效果

  Syntax-Aware Retrieval Augmented Code Generation 语法感知型检索增强代码生成

  A^3-CodGen: A Repository-Level Code Generation fr amework for Code Reuse with Local-Aware, Global-Aware, and Third-Party-Library-Aware A³-CodGen：兼顾本地、全局与第三方库的仓库级代码复用生成框架

  SkCoder: A Sketch-based Approach for Automatic Code Generation SkCoder：基于代码草图的自动代码生成方案

  CodeGen4Libs: A Two-Stage Approach for Library-Oriented Code Generation CodeGen4Libs：面向类库的两阶段代码生成方法

  ToolCoder: Teach Code Generation Models to use API search tools ToolCoder：训练代码生成模型调用API检索工具

  CodeAgent: Enhancing Code Generation with Tool-Integrated Agent Systems for Real-World Repo-level Coding Challenges CodeAgent：工具智能体助力解决仓库级代码开发难题

  RRGcode: Deep hierarchical search-based code generation RRGcode：基于深度分层检索的代码生成

  Code Search Is All You Need? Improving Code Suggestions with Code Search 代码检索赋能：依托检索优化代码推荐能力

  ARKS: Active Retrieval in Knowledge Soup for Code Generation ARKS：面向代码生成的知识混合池主动检索

• Code Summary 代码摘要

  Retrieval-based neural source code summarization 基于检索的神经源代码摘要生成

  Retrieve and Refine: Exemplar-based Neural Comment Generation 检索与优化：基于范例的神经代码注释生成

  EditSum: A Retrieve-and-Edit fr amework for Source Code Summarization EditSum：面向代码摘要的检索-编辑框架

  Retrieval-Augmented Generation for Code Summarization via Hybrid GNN 基于混合图神经网络的检索增强代码摘要生成

  Context-aware Retrieval-based Deep Commit Message Generation 上下文感知检索式深度提交日志生成

  RACE: Retrieval-augmented Commit Message Generation RACE：检索增强提交日志生成

  BashExplainer: Retrieval-Augmented Bash Code Comment Generation based on Fine-tuned CodeBERT BashExplainer：基于微调CodeBERT的检索增强Bash代码注释生成

  Retrieval-Based Transformer Pseudocode Generation 检索式Transformer伪代码生成

  A Simple Retrieval-based Method for Code Comment Generation 简易检索式代码注释生成方法

  READSUM: Retrieval-Augmented Adaptive Transformer for Source Code Summarization READSUM：检索增强自适应Transformer代码摘要模型

  Tram: A Token-level Retrieval-augmented Mechanism for Source Code Summarization Tram：词元级检索增强代码摘要机制

  Automatic Semantic Augmentation of Language Model Prompts (for Code Summarization) 面向代码摘要的语言模型提示语义自动增强

  Cross-Modal Retrieval-Enhanced Code Summarization based on Joint Learning for Retrieval and Generation 检生联合学习的跨模态检索增强代码摘要

  Automatic Smart Contract Comment Generation via Large Language Models and In-Context Learning 结合大模型与上下文学习的智能合约注释自动生成

  UniLog: Automatic Logging via LLM and In-Context Learning UniLog：依托大模型与上下文学习实现日志自动生成

• Code Completion 代码补全

  A Retrieve-and-Edit fr amework for Predicting Structured Outputs 面向结构化输出预测的检索-编辑框架

  Generating Code with the Help of Retrieved Template Functions and Stack Overflow Answers 借助检索模板函数与社区问答内容实现代码生成

  ReACC: A Retrieval-Augmented Code Completion fr amework ReACC：检索增强代码补全框架

  Domain Adaptive Code Completion via Language Models and Decoupled Domain Databases 结合语言模型与解耦领域库的领域自适应代码补全

  RepoCoder: Repository-Level Code Completion Through Iterative Retrieval and Generation RepoCoder：迭代检索与生成的仓库级代码补全

  CoCoMIC: Code Completion By Jointly Modeling In-file and Cross-file Context CoCoMIC：联合建模文件内与跨文件上下文的代码补全

  RepoFusion: Training Code Models to Understand Your Repository RepoFusion：让代码模型理解代码仓库的训练方案

  Revisiting and Improving Retrieval-Augmented Deep Assertion Generation 检索增强深度断言生成方法的优化与改进

  De-Hallucinator: Iterative Grounding for LLM-Based Code Completion De-Hallucinator：大模型代码补全的迭代事实锚定方案

  REPOFUSE: Repository-Level Code Completion with Fused Dual Context REPOFUSE：融合双重上下文的仓库级代码补全

• Automatic Program Repair 程序自动修复

  Repair Is Nearly Generation: Multilingual Program Repair with LLMs 修复即生成：基于大模型的多语言程序修复

  Retrieval-Based Prompt Selection for Code-Related Few-Shot Learning 代码类小样本学习的检索式提示筛选

  InferFix: End-to-End Program Repair with LLMs InferFix：基于大模型的端到端程序修复

  RAP-Gen: Retrieval-Augmented Patch Generation with CodeT5 for Automatic Program Repair RAP-Gen：结合CodeT5的检索增强补丁生成与程序自动修复

  Automated Code Editing with Search-Generate-Modify 检索-生成-修改流程实现代码自动编辑

  RTLFixer: Automatically Fixing RTL Syntax Errors with Large Language Models RTLFixer：依托大模型自动修复RTL语法错误

• Text-to-SQL and Code-based Semantic Parsing 文本转 SQL 与代码语义解析

  XRICL: Cross-lingual Retrieval-Augmented In-Context Learning for Cross-lingual Text-to-SQL Semantic Parsing 跨语言检索增强上下文学习用于跨语言Text-to-SQL解析

  Synchromesh: Reliable Code Generation from Pre-trained Language Models Synchromesh：基于预训练语言模型的高可靠代码生成

  Evaluating the Impact of Model Scale for Compositional Generalization in Semantic Parsing 模型规模对语义解析组合泛化能力的影响评估

  RESDSQL: Decoupling Schema Linking and Skeleton Parsing for Text-to-SQL RESDSQL：解耦表结构关联与骨架解析的Text-to-SQL方案

  Leveraging Code to Improve In-context Learning for Semantic Parsing 利用代码提升语义解析的上下文学习效果

  ReFSQL: A Retrieval-Augmentation fr amework for Text-to-SQL Generation ReFSQL：面向Text-to-SQL生成的检索增强框架

  Enhancing Text-to-SQL Capabilities of Large Language Models: A Study on Prompt Design Strategies 提升大模型Text-to-SQL能力：提示词设计策略研究

  Selective Demonstrations for Cross-domain Text-to-SQL 面向跨域Text-to-SQL的选择性样例引导

  DBCopilot: Scaling Natural Language Querying to Massive Databases via Schema Routing DBCopilot：依托表结构路由实现海量数据库自然语言查询

  Multi-Hop Table Retrieval for Open-Domain Text-to-SQL 面向开放域Text-to-SQL的多跳数据表检索

  CodeS: Towards Building Open-source Language Models for Text-to-SQL CodeS：面向Text-to-SQL任务的开源语言模型构建

• Others 其他

  De-fine: Decomposing and Refining Visual Programs with Auto-Feedback De-fine：结合自动反馈拆解并优化可视化程序

  Leveraging training data in few-shot prompting for numerical reasoning 小样本数值推理提示中训练数据的应用

  Retrieval-Augmented Code Generation for Universal Information Extraction 面向通用信息抽取的检索增强代码生成

  E&V: Prompting Large Language Models to Perform Static Analysis by Pseudo-code Execution and Verification E&V：通过伪代码执行与校验引导大模型完成静态分析

  Lessons from Building StackSpot AI: A Contextualized AI Coding Assistant StackSpot AI研发总结：具备上下文感知的AI编程助手

  Testing the Limits: Unusual Text Inputs Generation for Mobile App Crash Detection with Large Language Model 极限测试：利用大模型生成异常文本检测移动端应用崩溃

RAG for Audio 音频领域检索增强生成

• Audio Generation 音频生成

  Retrieval-Augmented Text-to-Audio Generation 检索增强文生音频

  Large-Scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation 特征融合与关键词增强的大规模语言-音频对比预训练

  Make-an-audio: Text-to-audio generation with prompt-enhanced diffusion models 提示增强扩散模型实现文生音频

• Audio Captioning 音频描述生成

  RECAP: Retrieval-Augmented Audio Captioning RECAP：检索增强音频描述生成

  Audio Captioning using Pre-Trained Large-Scale Language Model Guided by Audio-based Similar Caption Retrieval 相似音频描述检索引导预训练大语言模型完成音频字幕生成

  Large-Scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation 特征融合与关键词增强的大规模语言-音频对比预训练

  CNN architectures for large-scale audio classification 面向大规模音频分类的CNN网络结构

  Natural language supervision for general-purpose audio representations 通用音频表征的自然语言监督方案

  Weakly-supervised Automated Audio Captioning via text only training 仅基于文本训练的弱监督音频自动描述生成

  Training Audio Captioning Models without Audio 无音频数据训练音频描述生成模型

RAG for Image 图像领域检索增强生成

• Image Generation 图像生成

  Retrievegan: Image synthesis via differentiable patch retrieval RetrieveGAN：基于可微图像块检索的图像合成

  Instance-conditioned gan 实例条件生成对抗网络

  Memory-driven text-to-image generation 记忆驱动型文生图

  Re-imagen: Retrieval-augmented text-to-image generator Re-Imagen：检索增强文生图模型

  KNN-Diffusion: Image Generation via Large-Scale Retrieval KNN-Diffusion：基于大规模检索的图像生成

  Retrieval-Augmented Diffusion Models 检索增强扩散模型

  Text-Guided Synthesis of Artistic Images with Retrieval-Augmented Diffusion Models 检索增强扩散模型实现文本引导艺术图像生成

  X&Fuse: Fusing Visual Information in Text-to-Image Generation X&Fuse：文生图任务中的视觉信息融合方案

  Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs 多模态大模型赋能文生图扩散模型：重描述、规划与生成

• Image Captioning 图像描述生成

  Memory-augmented image captioning 记忆增强图像字幕生成

  Retrieval-enhanced adversarial training with dynamic memory-augmented attention for image paragraph captioning 动态记忆增强注意力与检索对抗训练实现图像段落描述

  Retrieval-Augmented Transformer for Image Captioning 面向图像字幕生成的检索增强Transformer

  Retrieval-augmented image captioning 检索增强图像字幕生成

  Reveal: Retrieval-augmented visual-language pre-training with multi-source multimodal knowledge memory Reveal：多源多模态知识记忆的检索增强视觉-语言预训练

  SmallCap: Lightweight Image Captioning Prompted With Retrieval Augmentation SmallCap：检索增强提示驱动的轻量图像字幕模型

  Cross-Modal Retrieval and Semantic Refinement for Remote Sensing Image Captioning 面向遥感图像描述的跨模态检索与语义优化

• Others 其他

  An empirical study of gpt-3 for few-shot knowledge-based vqa GPT-3用于小样本知识型视觉问答的实证研究

  Retrieval augmented visual question answering with outside knowledge 结合外部知识的检索增强视觉问答

  Augmenting transformers with KNN-based composite memory for dialog 近邻复合记忆增强Transformer用于对话任务

  Maria: A visual experience powered conversational agent Maria：融合视觉信息的对话智能体

  Neural machine translation with phrase-level universal visual representations 融合短语级通用视觉表征的神经机器翻译

RAG for Video 视频领域检索增强生成

• Video Captioning 视频描述生成

  Incorporating Background Knowledge into Video Desc ription Generation 将背景知识融入视频描述生成

  Retrieval Augmented Convolutional Encoder-decoder Networks for Video Captioning 检索增强卷积编解码网络实现视频描述生成

  Concept-Aware Video Captioning: Describing Videos With Effective Prior Information 概念感知视频描述：利用有效先验信息生成视频文案

  Retrieval-Augmented Egocentric Video Captioning 检索增强第一视角视频描述生成

• Video QA&Dialogue 视频问答与对话

  Memory augmented deep recurrent neural network for video question answering 记忆增强深度循环网络用于视频问答

  Retrieving-to-answer: Zero-shot video question answering with frozen large language models 检索作答：基于冻结大模型的零样本视频问答

  Tvqa+: Spatio-temporal grounding for video question answering TVQA+：面向视频问答的时空信息锚定

  Vgnmn: Video-grounded neural module networks for video-grounded dialogue systems VGNMN：视频感知神经模块网络用于视频对话系统

• Others 其他

  Language models with image desc riptors are strong few-shot video-language learners 搭载图像描述器的语言模型具备优秀的小样本视频-语言学习能力

  RAG-Driver: Generalisable Driving Explanations with Retrieval-Augmented In-Context Learning in Multi-Modal Large Language Model RAG-Driver：多模态大模型检索增强上下文学习实现通用驾驶行为解读

  Animate-A-Story: Storytelling with Retrieval-Augmented Video Generation Animate-A-Story：检索增强视频生成助力故事创作

  Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval 时空冻结：面向端到端检索的视频图像联合编码器

RAG for 3D 三维领域检索增强生成

• Text-to-3D 文生三维模型

  ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model ReMoDiffuse：检索增强动作扩散模型

  AMD: Anatomical Motion Diffusion with Interpretable Motion Decomposition and Fusion AMD：具备可解释动作拆解与融合的人体动作扩散模型

  Retrieval-Augmented Score Distillation for Text-to-3D Generation 检索增强分数蒸馏用于文生三维模型

RAG for Knowledge 知识领域检索增强生成

• Knowledge Base Question Answering 知识库问答

  ReTraCk: A Flexible and Efficient fr amework for Knowledge Base Question Answering ReTraCk：灵活高效的知识库问答框架

  Unseen Entity Handling in Complex Question Answering over Knowledge Base via Language Generation 借助语言生成处理知识库复杂问答中的未知实体

  Case-based Reasoning for Natural Language Queries over Knowledge Bases 面向知识库自然语言查询的案例推理

  Logical Form Generation via Multi-task Learning for Complex Question Answering over Knowledge Bases 基于多任务学习的知识库复杂问答逻辑表达式生成

  Uni-Parser: Unified Semantic Parser for Question Answering on Knowledge Base and Database Uni-Parser：适配知识库与数据库问答的统一语义解析器

  RNG-KBQA: Generation Augmented Iterative Ranking for Knowledge Base Question Answering RNG-KBQA：生成增强迭代排序的知识库问答方案

  TIARA: Multi-grained Retrieval for Robust Question Answering over Large Knowledge Base TIARA：面向大规模知识库问答的多粒度检索

  DecAF: Joint Decoding of Answers and Logical Forms for Question Answering over Knowledge Bases DecAF：知识库问答的答案与逻辑表达式联合解码

  End-to-end Case-Based Reasoning for Commonsense Knowledge Base Completion 面向常识知识库补全的端到端案例推理

  Bridging the KB-Text Gap: Leveraging Structured Knowledge-aware Pre-training for KBQA 弥合知识库与文本鸿沟：结构化知识预训练赋能知识库问答

  Knowledge-Driven CoT: Exploring Faithful Reasoning in LLMs for Knowledge-intensive Question Answering 知识驱动思维链：探究大模型在知识密集型问答中的可信推理

  Few-shot Transfer Learning for Knowledge Base Question Answering: Fusing Supervised Models with In-Context Learning 知识库问答小样本迁移学习：融合监督模型与上下文学习

  FC-KBQA: A Fine-to-Coarse Composition fr amework for Knowledge Base Question Answering FC-KBQA：由细到粗组合式知识库问答框架

  Knowledge-Augmented Language Model Prompting for Zero-Shot Knowledge Graph Question Answering 面向零样本知识图谱问答的知识增强提示方法

  Knowledge Graph-augmented Language Models for Complex Question Answering 知识图谱增强语言模型实现复杂问答

  Retrieve-Rewrite-Answer: A KG-to-Text Enhanced LLMs fr amework for Knowledge Graph Question Answering 检索-改写-作答：知识图谱转文本增强的大模型问答框架

  Distribution Shifts Are Bottlenecks: Extensive Evaluation for Grounding Language Models to Knowledge Bases 分布偏移问题研究：大模型对接知识库的综合评估

  Probing Structured Semantics Understanding and Generation of Language Models via Question Answering 借助问答任务探测语言模型的结构化语义理解与生成能力

  Keqing: Knowledge-based Question Answering is A Nature Chain-of-Thought mentor of LLMs Keqing：知识库问答作为大模型天然的思维链引导范式

  Interactive-KBQA: Multi-Turn Interactions for Knowledge Base Question Answering with Large Language Models Interactive-KBQA：基于大模型的多轮交互式知识库问答

• Knowledge-augmented Open-domain Question Answering 知识增强开放域问答

  UniK-QA: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering UniK-QA：融合结构化与非结构化知识表征的开放域问答

  KG-FiD: Infusing Knowledge Graph in Fusion-in-Decoder for Open-Domain Question Answering KG-FiD：知识图谱融入解码器融合模型实现开放域问答

  Empowering Language Models with Knowledge Graph Reasoning for Open-Domain Question Answering 知识图谱推理增强语言模型用于开放域问答

  Grape: Knowledge Graph Enhanced Passage Reader for Open-domain Question Answering Grape：知识图谱增强文本检索器实现开放域问答

  Enhancing Multi-modal Multi-hop Question Answering via Structured Knowledge and Unified Retrieval-Generation 结构化知识与统一检生框架增强多模态多跳问答

  DIVKNOWQA: Assessing the Reasoning Ability of LLMs via Open-Domain Question Answering over Knowledge Base and Text DIVKNOWQA：基于知识库与文本混合开放域问答评估大模型推理能力

  KnowledGPT: Enhancing Large Language Models with Retrieval and Storage Access on Knowledge Bases KnowledGPT：依托知识库检索与存储能力增强大模型

  Evidence-Focused Fact Summarization for Knowledge-Augmented Zero-Shot Question Answering 面向知识增强零样本问答的证据导向事实摘要

  Two-stage Generative Question Answering on Temporal Knowledge Graph Using Large Language Models 基于大模型的时序知识图谱两阶段生成式问答

  KnowledgeNavigator: Leveraging Large Language Models for Enhanced Reasoning over Knowledge Graph KnowledgeNavigator：利用大模型强化知识图谱推理能力

  GNN-RAG: Graph Neural Retrieval for Large Language Model Reasoning GNN-RAG：图神经检索助力大模型推理

• Table Question Answering 表格问答

  NeurIPS 2020 EfficientQA Competition: Systems, Analyses and Lessons Learned NeurIPS 2020高效问答竞赛：参赛方案、分析与经验总结

  Dual Reader-Parser on Hybrid Textual and Tabular Evidence for Open Domain Question Answering 面向混合文本与表格数据的双路检索解析开放域问答

  End-to-End Table Question Answering via Retrieval-Augmented Generation 基于检索增强生成的端到端表格问答

  OmniTab: Pretraining with Natural and Synthetic Data for Few-shot Table-based Question Answering OmniTab：结合真实与合成数据预训练的小样本表格问答模型

  Reasoning over Hybrid Chain for Table-and-Text Open Domain Question Answering 面向表格与文本混合数据的链式推理开放域问答

  Conversational Question Answering on Heterogeneous Sources 基于多源异构数据的对话式问答

  Open-domain Question Answering via Chain of Reasoning over Heterogeneous Knowledge 异构知识链式推理实现开放域问答

  StructGPT: A General fr amework for Large Language Model to Reason over Structured Data StructGPT：大模型面向结构化数据推理的通用框架

  cTBLS: Augmenting Large Language Models with Conversational Tables cTBLS：对话式表格增强大语言模型

  RINK: Reader-Inherited Evidence Reranker for Table-and-Text Open Domain Question Answering RINK：继承检索特征的证据重排序模型用于表格文本混合问答

  Localize, Retrieve and Fuse: A Generalized fr amework for Free-Form Question Answering over Tables 定位-检索-融合：通用自由式表格问答框架

  Exploring the Impact of Table-to-Text Methods on Augmenting LLM-based Question Answering with Domain Hybrid Data 表格转文本方法对领域混合数据增强大模型问答的影响探究

  ERATTA: Extreme RAG for Table To Answers with Large Language Models ERATTA：面向表格问答的极致检索增强生成方案

• Others 其他

  Improving Knowledge-Aware Dialogue Response Generation by Using Human-Written Prototype Dialogues 利用人工范例对话优化知识感知型回复生成

  Knowledge Graph-Augmented Language Models for Knowledge-Grounded Dialogue Generation 知识图谱增强语言模型实现知识锚定对话生成

  RHO: Reducing Hallucination in Open-domain Dialogues with Knowledge Grounding RHO：依托知识锚定缓解开放域对话幻觉问题

  Retrieval-Enhanced Generative Model for Large-Scale Knowledge Graph Completion 检索增强生成模型用于大规模知识图谱补全

  Knowledge-Augmented Large Language Models for Personalized Contextual Query Suggestion 知识增强大模型实现个性化上下文查询推荐

  G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering G-Retriever：面向文本图理解与问答的检索增强生成

  RAG-based Explainable Prediction of Road Users Behaviors for Automated Driving using Knowledge Graphs and Large Language Models 结合知识图谱与大模型的检索增强自动驾驶行为可解释预测

  HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models HippoRAG：受神经生物学启发的大模型长期记忆检索增强方案

RAG for Science 科学领域检索增强生成

• Drug Discovery 药物研发

  Retrieval-based controllable molecule generation 基于检索的可控分子生成

  Prompt-based 3d molecular diffusion models for structure-based drug design 基于提示的三维分子扩散模型用于靶点药物设计

• Biomedical Informatics Enhancement 生物医学信息优化

  PoET: A generative model of protein families as sequences-of-sequences PoET：面向蛋白质家族的序列嵌套生成模型

  Retrieval-augmented large language models for adolescent idiopathic scoliosis patients in shared decision-making 检索增强大模型用于青少年特发性脊柱侧弯诊疗协同决策

  BioReader: a Retrieval-Enhanced Text-to-Text Transformer for Biomedical Literature BioReader：检索增强文本转换Transformer用于生物医学文献处理

  Writing by Memorizing: Hierarchical Retrieval-based Medical Report Generation 记忆式写作：分层检索医疗报告生成方案

  From RAG to QA-RAG: Integrating Generative AI for Pharmaceutical Regulatory Compliance Process 从RAG到QA-RAG：生成式AI赋能医药合规审查流程

  RAG-RLRC-LaySum at BioLaySumm: Integrating Retrieval-Augmented Generation and Readability Control for Layman Summarization of Biomedical Texts RAG-RLRC-LaySum：融合检索增强生成与可读性控制的生物医学文献通俗摘要

• Math Applications 数学应用

  Retrieval-augmented Generation to Improve Math Question-Answering: Trade-offs Between Groundedness and Human Preference 检索增强生成优化数学问答：事实可信度与人工偏好的权衡

  LeanDojo: Theorem Proving with Retrieval-Augmented Language Models LeanDojo：基于检索增强语言模型的定理证明

Benchmark 评测基准

Benchmarking Large Language Models in Retrieval-Augmented Generation 检索增强生成中大语言模型的基准评测

CRUD-RAG: A Comprehensive Chinese Benchmark for Retrieval-Augmented Generation of Large Language Models CRUD-RAG：面向大语言模型检索增强生成的中文综合评测基准

ARES: An Automated Evaluation fr amework for Retrieval-Augmented Generation Systems ARES：检索增强生成系统自动化评估框架

RAGAS: Automated Evaluation of Retrieval Augmented Generation RAGAS：检索增强生成自动化评测工具

KILT: a Benchmark for Knowledge Intensive Language Tasks KILT：面向知识密集型语言任务的评测基准

Citation

if you find this work useful, please cite our paper: 若本内容对你有所帮助，请引用我们的论文：

@article{zhao2024retrieval,
  title={Retrieval-Augmented Generation for AI-Generated Content: A Survey},
  author={Zhao, Penghao and Zhang, Hailin and Yu, Qinhan and Wang, Zhengren and Geng, Yunteng and Fu, Fangcheng and Yang, Ling and Zhang, Wentao and Cui, Bin},
  journal={arXiv preprint arXiv:2402.19473},
  year={2024}
}

论文译文
标题：面向人工智能内容生成的检索增强生成技术综述
作者：赵鹏浩、张翰林、余钦翰、王振仁、耿云腾、付方成、杨玲、张文韬、崔斌
期刊：arXiv预印本 arXiv:2402.19473
年份：2024

推荐学习书籍 《CDA一级教材》适合CDA一级考生备考，也适合业务及数据分析岗位的从业者提升自我。完整电子版已上线CDA网校，累计已有10万+在读~ !

免费加入阅读：https://edu.cda.cn/goods/show/3151?targetId=5147&preview=0