Active Retrieval Augmented Generation
2023/05/11 by Zhengbao Jiang, Jiang, Zhengbao, Frank F. Xu +15 · 70 citations
Computer Science · #Topic Modeling #Natural Language Processing Techniques #Multimodal Machine Learning Applications
paper · pdf · doi:10.48550/arxiv.2305.06983
Abstract
Despite the remarkable ability of large language models (LMs) to comprehend and generate language, they have a tendency to hallucinate and create factually inaccurate output. Augmenting LMs by retrieving information from external knowledge resources is one promising solution. Most existing retrieval augmented LMs employ a retrieve-and-generate setup that only retrieves information once based on the input. This is limiting, however, in more general scenarios involving generation of long texts, where continually gathering information throughout generation is essential. In this work, we provide a generalized view of active retrieval augmented generation, methods that actively decide when and what to retrieve across the course of the generation. We propose Forward-Looking Active REtrieval augmented generation (FLARE), a generic method which iteratively uses a prediction of the upcoming sentence to anticipate future content, which is then utilized as a query to retrieve relevant documents to regenerate the sentence if it contains low-confidence tokens. We test FLARE along with baselines comprehensively over 4 long-form knowledge-intensive generation tasks/datasets. FLARE achieves superior or competitive performance on all tasks, demonstrating the effectiveness of our method. Code and datasets are available at https://github.com/jzbjyb/FLARE.
Cited by
- Bidirectional RAG: Safe Self-Improving Retrieval-Augmented Generation Through Multi-Stage Validation
- MedJudgeRAG: Option-Wise Evidence Judgment with Dynamic Knowledge Graphs for Medical MCQA
- Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation
- Structure Over Scale: Schema-Constrained Causal Graphs for RAG
- SF-AMS: Strategic Forgetting for Structured Memory in LLM Agent
- Procedural Knowledge at Scale Improves Reasoning
- Graph-based Nearest Neighbors with Dynamic Updates via Random Walks
- AdaSearch: Balancing Parametric Knowledge and Search in Large Language Models via Reinforcement Learning
- Replace, Don't Expand: Mitigating Context Dilution in Multi-Hop RAG via Fixed-Budget Evidence Assembly
- KBQA-R1: Reinforcing Large Language Models for Knowledge Base Question Answering
- Living the Novel: A System for Generating Self-Training Timeline-Aware Conversational Agents from Novels
- The Road of Adaptive AI for Precision in Cybersecurity
- WISE: Weighted Iterative Society-of-Experts for Robust Multimodal Multi-Agent Debate
- EmoRAG: Evaluating RAG Robustness to Symbolic Perturbations
- RAGSmith: A Framework for Finding the Optimal Composition of Retrieval-Augmented Generation Methods Across Datasets
- BudgetLeak: Membership Inference Attacks on RAG Systems via the Generation Budget Side Channel
- Retrieval as a Decision: Training-Free Adaptive Gating for Efficient RAG
- MemSearcher: Training LLMs to Reason, Search and Manage Memory via End-to-End Reinforcement Learning
- Beyond Single Embeddings: Capturing Diverse Targets with Multi-Query Retrieval
- Interact-RAG: Reason and Interact with the Corpus, Beyond Black-Box Retrieval
- InfoFlow: Reinforcing Search Agent Via Reward Density Optimization
- Optimizing Retrieval for RAG via Reinforced Contrastive Learning
- Reinforcement Learning for Long-Horizon Multi-Turn Search Agents
- A Survey of Data Agents: Emerging Paradigm or Overstated Hype?
- AcademicEval: Live Long-Context LLM Benchmark
- Stop-RAG: Value-Based Retrieval Control for Iterative RAG
- A survey on retrieval-augmentation generation (RAG) models for healthcare applications
- Toward Cybersecurity-Expert Small Language Models
- Learnable Game-theoretic Policy Optimization for Data-centric Self-explanation Rationalization
- Beyond Correctness: Rewarding Faithful Reasoning in Retrieval-Augmented Generation
- Uncertainty Quantification for Hallucination Detection in Large Language Models: Foundations, Methodology, and Future Directions
- SAIL-Embedding Technical Report: Omni-modal Embedding Foundation Model
- Query-Specific GNN: A Comprehensive Graph Representation Learning Method for Retrieval Augmented Generation
- Uncertainty Quantification for Retrieval-Augmented Reasoning
- AssoMem: Scalable Memory QA with Multi-Signal Associative Retrieval
- CardRewriter: Leveraging Knowledge Cards for Long-Tail Query Rewriting on Short-Video Platforms
- LinearRAG: Linear Graph Retrieval Augmented Generation on Large-scale Corpora
- When Retrieval Succeeds and Fails: Rethinking Retrieval-Augmented Generation for LLMs
- Haystack Engineering: Context Engineering for Heterogeneous and Agentic Long-Context Evaluation
- MASH: Modeling Abstention via Selective Help-Seeking
- ReSeek: A Self-Correcting Framework for Search Agents with Instructive Rewards
- Rationale-Augmented Retrieval with Constrained LLM Re-Ranking for Task Discovery
- Emission-GPT: A domain-specific language model agent for knowledge retrieval, emission inventory and data analysis
- GeoEvolve: Automating Geospatial Model Discovery via Multi-Agent Large Language Models
- Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoning
- Dr-DCI: Scaling Direct Corpus Interaction via Dynamic Workspace Expansion
- AgentClinic: a multimodal benchmark for tool-using clinical AI agents
- A State-Update Prompting Strategy for Efficient and Robust Multi-turn Dialogue
- Do Large Language Models Favor Recent Content? A Study on Recency Bias in LLM-Based Reranking
- A Survey on Retrieval And Structuring Augmented Generation with Large Language Models
- Deploying AI for Signal Processing education: Selected challenges and intriguing opportunities
- PromptGuard: An Orchestrated Prompting Framework for Principled Synthetic Text Generation for Vulnerable Populations using LLMs with Enhanced Safety, Fairness, and Controllability
- Exploratory Retrieval-Augmented Planning For Continual Embodied Instruction Following
- VerlTool: Towards Holistic Agentic Reinforcement Learning with Tool Use
- EviNote-RAG: Enhancing RAG Models via Answer-Supportive Evidence Notes
- SHERPA: A Model-Driven Framework for Large Language Model Execution
- AI-SearchPlanner: Modular Agentic Search via Pareto-Optimal Multi-Objective Reinforcement Learning
- LFD: Layer Fused Decoding to Exploit External Knowledge in Retrieval-Augmented Generation
- Retrieval-Augmented Generation for Natural Language Art Provenance Searches in the Getty Provenance Index
- VistaWise: Building Cost-Effective Agent with Cross-Modal Knowledge Graph for Minecraft
- AdaDocVQA: Adaptive Framework for Long Document Visual Question Answering in Low-Resource Settings
- FLAIR: Feedback Learning for Adaptive Information Retrieval
- Fast, Slow, and Tool-augmented Thinking for LLMs: A Review
- Towards Multimodal Sentiment Analysis via Contrastive Cross-modal Retrieval Augmentation and Hierachical Prompts
- PrLM: Learning Explicit Reasoning for Personalized RAG via Contrastive Reward Optimization
- Synthesizing scientific literature with retrieval-augmented language models
- Impact-driven Context Filtering For Cross-file Code Completion
- LMAR: Language Model Augmented Retriever for Domain-specific Knowledge Indexing
- Thinking Machines: Mathematical Reasoning in the Age of LLMs
- Lucy: edgerunning agentic web search on mobile with machine generated task vectors
Related