EviNote-RAG: Enhancing RAG Models via Answer-Supportive Evidence Notes

2025/08/31 by Dai, Yuqin, Guoqing Wang, Yuan Wang +28 · 5 citations
Computer Science · #Computation and Language (cs.CL) #Expert finding and Q&A systems #FOS: Computer and information sciences #Multimodal Machine Learning Applications #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2509.00877

openalex publication_date 2025/08/31 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Retrieval-Augmented Generation (RAG) has advanced open-domain question answering by incorporating external information into model reasoning. However, effectively leveraging external information to enhance reasoning presents the following challenges: (1) low signal-to-noise ratio, where answer-supportive external information is diluted by irrelevant material, and (2) error accumulation, which arises in multi-hop reasoning when incomplete or misleading information is incorporated. To address these challenges, we introduce EviNote-RAG, a framework that follows a retrieve-note-answer workflow. Instead of reasoning directly over raw external information, the model first produces Supportive-Evidence Notes (SENs), which concisely preserve answer-critical information and explicitly mark key and uncertainty information to improve accuracy. We further design an entailment-based Evidence Quality Reward (EQR) to ensure that SENs are logically sufficient to derive the final answer, thereby enhancing SENs' quality. Experiments on both in-domain and out-of-domain QA benchmarks show that EviNote-RAG achieves state-of-the-art performance, improving answer accuracy, training stability, robustness, and efficiency. In particular, it yields relative F1 gains of 20% on HotpotQA (+0.093), 40% on Bamboogle (+0.151), and 91% on 2Wiki (+0.256), benefiting from improvements in the reasoning process.

Citations

• Atom-Searcher: Enhancing Agentic Deep Research via Fine-Grained Atomic Thought Reward
• Careful Queries, Credible Results: Teaching RAG Models Advanced Web Search Tools with Reinforcement Learning
• R-Search: Empowering LLM Reasoning with Search via Multi-Reward Reinforcement Learning
• R1-Searcher++: Incentivizing the Dynamic Knowledge Acquisition of LLMs via Reinforcement Learning
• WebAgent-R1: Training Web Agents via End-to-End Multi-Turn Reinforcement Learning
• ZeroSearch: Incentivize the Search Capability of LLMs without Searching
• ReasonIR: Training Retrievers for Reasoning Tasks
• Synergizing RAG and Reasoning: A Systematic Review
• DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environments
• ReaRAG: Knowledge-guided Reasoning Enhances Factuality of Large Reasoning Models with Iterative Retrieval Augmented Generation
• Open Deep Search: Democratizing Search with Open-source Reasoning Agents
• RAG-RL: Advancing Retrieval-Augmented Generation via RL and Curriculum Learning
• Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
• Beyond Outlining: Heterogeneous Recursive Planning for Adaptive Long-form Writing with Language Models
• R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning
• From RAG to Memory: Non-Parametric Continual Learning for Large Language Models
• Agentic Reasoning: A Streamlined Framework for Enhancing LLM Reasoning with Agentic Tools
• MedRAG: Enhancing Retrieval-augmented Generation with Knowledge Graph-Elicited Reasoning for Healthcare Copilot
• DeepRAG: Thinking to Retrieve Step by Step for Large Language Models
• DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning
• PIKE-RAG: sPecIalized KnowledgE and Rationale Augmented Generation
• ReARTeR: Retrieval-Augmented Reasoning with Trustworthy Process Rewarding
• Search-o1: Agentic Search-Enhanced Large Reasoning Models
• Qwen2.5 Technical Report
• An Agent Framework for Real-Time Financial Information Searching with Large Language Models
• RARE: Retrieval-Augmented Reasoning Enhancement for Large Language Models
• WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning
• Plan*RAG: Efficient Test-Time Planning for Retrieval Augmented Generation
• Long-Context LLMs Meet RAG: Overcoming Challenges for Long Inputs in RAG
• Inference Scaling for Long-Context Retrieval Augmented Generation
• Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks
• A Compressive Memory-based Retrieval Approach for Event Argument Extraction
• Into the Unknown Unknowns: Engaged Human Learning through Participation in Language Model Agent Conversations
• Hierarchical Retrieval-Augmented Generation Model with Rethink for Multi-hop Question Answering
• Summary of a Haystack: A Challenge to Long-Context LLMs and RAG Systems
• PlanRAG: A Plan-then-Retrieval Augmented Generation for Generative Large Language Models as Decision Makers
• RE-RAG: Improving Open-Domain QA Performance and Interpretability with Relevance Estimator in Retrieval-Augmented Generation
• Chain of Agents: Large Language Models Collaborating on Long-Context Tasks
• Don't Forget to Connect! Improving RAG with Graph-based Reranking
• SuRe: Summarizing Retrievals using Answer Candidates for Open-domain QA of LLMs
• Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity
• RAFT: Adapting Language Model to Domain Specific RAG
• Metacognitive Retrieval-Augmented Large Language Models
• BioXP-0.5B: Explainable Medical-AI via RL-GRPO
• Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection
• LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression
• RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentation
• Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models
• 🧜Siren’s Song in the AI Ocean: A Survey on Hallucination in Large Language Models
• Ghostbuster: Detecting Text Ghostwritten by Large Language Models
• Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergy
• Active Retrieval Augmented Generation
• Large Language Models Can Be Easily Distracted by Irrelevant Context
• Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions
• Text Embeddings by Weakly-Supervised Contrastive Pre-training
• Query Refinement Prompts for Closed-Book Long-Form Question Answering
• Measuring and Narrowing the Compositionality Gap in Language Models
• Distributing Accountability, Not Capability: Phase Separation and the LLM Workflow Quadrant in Autonomous AI Agent Architectures
• Re2G: Retrieve, Rerank, Generate
• PaLM: Scaling Language Modeling with Pathways
• Survey of Hallucination in Natural Language Generation
• WebGPT: Browser-assisted question-answering with human feedback
• Evidentiality-guided Generation for Knowledge-Intensive NLP Tasks
• MuSiQue: Multihop Questions via Single-hop Question Composition
• Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Steps
• Language Models are Few-Shot Learners
• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• Dense Passage Retrieval for Open-Domain Question Answering
• REALM: Retrieval-Augmented Language Model Pre-Training
• DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
• HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering
• TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for\n Reading Comprehension
• Retrieval is Not Enough: Enhancing RAG Reasoning through Test-Time Critique and Optimization
• Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Models
• When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories

Cited by

• Information Gain-based Policy Optimization: A Simple and Effective Approach for Multi-Turn Search Agents
• Code-driven Number Sequence Calculation: Enhancing the inductive Reasoning Abilities of Large Language Models
• NCV: A Node-Wise Consistency Verification Approach for Low-Cost Structured Error Localization in LLM Reasoning
• Query Disambiguation via Answer-Free Context: Doubling Performance on Humanity's Last Exam
• Contextual Information Policy Optimization for Search Agents
• Consensus Entropy: Harnessing Multi-VLM Agreement for Self-Verifying and Self-Improving OCR

Related

• TRACE the Evidence: Constructing Knowledge-Grounded Reasoning Chains for Retrieval-Augmented Generation
• Stateful Evidence-Driven Retrieval-Augmented Generation with Iterative Reasoning
• Learning to Extract Rational Evidence via Reinforcement Learning for Retrieval-Augmented Generation
• SURE-RAG: Sufficiency and Uncertainty-Aware Evidence Verification for Selective Retrieval-Augmented Generation
• IAG: Induction-Augmented Generation Framework for Answering Reasoning Questions