mlsanityData to 14 Oct 2026

Bridging Relevance and Reasoning: Rationale Distillation in Retrieval-Augmented Generation

arXiv 2412.08519 ↗First posted 11 Dec 2024cs.CLcited by 13

On the map: LLM reasoning → Knowledge- and tool-augmented reasoning

also in 1 more
Abstract (arXiv)

The reranker and generator are two critical components in the Retrieval-Augmented Generation (i.e., RAG) pipeline, responsible for ranking relevant documents and generating responses. However, due to differences in pre-training data and objectives, there is an inevitable gap between the documents ranked as relevant by the reranker and those required by the generator to support answering the query. To address this gap, we propose RADIO, a novel and practical preference alignment framework with RAtionale DIstillatiOn. Specifically, we first propose a rationale extraction method that leverages the reasoning capabilities of Large Language Models (LLMs) to extract the rationales necessary for answering the query. Subsequently, a rationale-based alignment process is designed to rerank the documents based on the extracted rationales, and fine-tune the reranker to align the preferences. We conduct extensive experiments on two tasks across three datasets to demonstrate the effectiveness of our approach compared to baseline methods. Our code is released online to ease reproduction.

Key facts. Cited by 13 papers in the map (data through 14 Oct 2026). Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning. Builds on 26 of its 46 references that are in the map. First posted on arXiv 11 Dec 2024.

Builds on

26 of 46 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Retrieval-Augmented Generation for Large Language Models: A Survey18 Dec

Other references · 25

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  3. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  4. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  5. 2020Scaling Laws for Neural Language Modelscited by 6,586
  6. 2023A Survey of Large Language Modelscited by 2,576
  7. 2024Qwen2 Technical Reportcited by 2,120
  8. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,253
  9. 2024M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillationcited by 1,073
  10. 2023REPLUG: Retrieval-Augmented Black-Box Language Modelscited by 792
  11. 2023Towards General Text Embeddings with Multi-stage Contrastive Learningcited by 723
  12. 2023C-Pack: Packed Resources For General Chinese Embeddingscited by 578
13 more
  1. 2023Query Rewriting for Retrieval-Augmented Large Language Modelscited by 213
  2. 2024FlashRAG: A Modular Toolkit for Efficient Retrieval-Augmented Generation Researchcited by 166
  3. 2024NV-Retriever: Improving text embedding models with effective hard-negative miningcited by 84
  4. 2023Augmentation-Adapted Retriever Improves Generalization of Language Models as Generic Plug-Incited by 71
  5. 2024Bridging the Preference Gap between Retrievers and LLMscited by 69
  6. 2024Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented Generationcited by 27
  7. 2024Large Language Model Distilling Medication Recommendation Modelcited by 26
  8. 2024A Survey on Human Preference Learning for Large Language Modelscited by 24
  9. 2024Multi-perspective Improvement of Knowledge Graph Completion with Large Language Modelscited by 20
  10. 2023Agent4Ranking: Semantic Robust Ranking via Personalized Query Rewriting Using Multi-agent LLMcited by 18
  11. 2024Enhancing Q&A Text Retrieval with Ranking Models: Benchmarking, fine-tuning and deploying Rerankers for RAGcited by 16
  12. 2024ARL2: Aligning Retrievers for Black-box Large Language Models via Self-guided Adaptive Relevance Labelingcited by 10
  13. 2023Large Multimodal Model Compression via Efficient Pruning and Distillation at AntGroupcited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 9; 2026: 4.

13 papers.

Sort:

By area:

  1. 2025Process vs. Outcome Reward: Which is Better for Agentic RAG Reinforcement Learningcited by 70
  2. 2025A Survey of Personalization: From RAG to Agentcited by 27
  3. 2025Ranking Free RAG: Replacing Re-ranking with Selection in RAG for Sensitive Domainscited by 5
  4. 2026MemSearch-o1: Empowering Large Language Models with Reasoning-Aligned Memory Growth in Agentic Searchcited by 5
  5. 2025TAPO: Task-Referenced Adaptation for Prompt Optimizationcited by 3
  6. 2025HetaRAG: Hybrid Deep Retrieval-Augmented Generation across Heterogeneous Data Storescited by 3
  7. 2025The Evolution of Reranking Models in Information Retrieval: From Heuristic Methods to Large Language Modelscited by 3
  8. 2025ZeroED: Hybrid Zero-shot Error Detection through Large Language Model Reasoningcited by 2
  9. 2025TabRAG: Improving Tabular Document Question Answering for Retrieval Augmented Generation via Structured Representationscited by 1
  10. 2026Knowledge Distillation Must Account for What It Losescited by 1
  11. 2025Rethinking LLM Parametric Knowledge as Post-retrieval Confidence for Dynamic Retrieval and Rerankingcited by 0
  12. 2026When Hard Negatives Hurt: Bridging the Generative-Discriminative Gap in Hard Negative Synthesis for Retrievalcited by 0
1 more
  1. 2026Re-Ranking Through an Attribution Lens for Citation Quality in Legal QAcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Rethinking Retrieval-Augmented Generation as a Cooperative Decision-Making Problem
  2. 2025LIR3AG: A Lightweight Rerank Reasoning Strategy Framework for Retrieval-Augmented Generation
  3. 2025Aligning Extraction and Generation for Robust Retrieval-Augmented Generation
  4. 2025Distilling a Small Utility-Based Passage Selector to Enhance Retrieval-Augmented Generation
  5. 2024JudgeRank: Leveraging Large Language Models for Reasoning-Intensive Reranking
  6. 2026Less is More for RAG: Information Gain Pruning for Generator-Aligned Reranking and Evidence Selection
  7. 2026Efficient Rationale-based Retrieval: On-policy Distillation from Generative Rerankers based on JEPA
  8. 2025Enhancing Retrieval-Augmented Generation with Two-Stage Retrieval: FlashRank Reranking and Query Expansion