Aug 2026· Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2· pp. 13308-13313· 0 citations· 27 references
Abstract
Retrieval-Augmented Generation (RAG) has emerged as the dominant strategy to ground LLM outputs in retrieved evidence to reduce hallucinations and increase factuality, yet RAG itself introduces new challenges: noisy retrieval, knowledge conflicts, and wasted computation from unnecessary retrieval. This tutorial provides an in-depth treatment of modern RAG based on AI-facilitated systematic analysis of ~2000 recent papers (2020--2026). We trace the RAG pipeline from its modular foundations through graph-enhanced reasoning to the latest RL-driven agentic architectures, covering each stage---triggering, query rewriting, retrieval, post-processing, generation, and knowledge-graph integration---with running examples, method comparisons, and benchmark results. Attendees will leave with a structured roadmap, evidence-backed practical recommendations, and a clear map of open research opportunities.
Retrieval-augmented generation (RAG) enhances large language models (LLMs) by incorporating external information, but traditional fixed retrieval processes struggle to adapt to complex task requirements. In recent years, reinforcement learning (RL) has been increasingly applied to train LLMs to autonomously invoke search tools, driving RAG to evolve from the passive information acquisition of a fixed pipeline to a trustworthy retrieval system with autonomous decision-making capabilities. This paper reviews the representative studies on the combination of LLMs, RAG and RL in recent years. It focuses on analyzing the role of RL in dynamic retrieval, process rewards, query optimization, etc., and compares the connections and evolutionary relationships among different methods. The research findings show that RL has gradually expanded from simply improving the accuracy of the final answer to optimizing queries, multi-round search, process decision-making and trustworthy screening, providing new ideas for enhancing the active retrieval ability of RAG and improving the credibility of information.
Zun-Long Hong· Applied and Computational En...· 0 citations
Retrieval-Augmented Generation (RAG) has become a fundamental paradigm for enhancing Large Language Models (LLMs) with external knowledge. However, while recent structure-augmented approaches organize documents into graphs to improve information access, their retrieval strategies remain largely static, relying on similarity ranking or static probability diffusion. We identify that this paradigm suffers from two inherent limitations in complex reasoning: popularity bias, where retrieval paths are trapped by high-degree distractors, and signal decay, where relevance signals attenuate over long reasoning chains. To overcome these challenges, we propose NaviRAG, a novel framework that reformulates retrieval as a reinforcement learning-driven dynamic navigation problem on schema-less knowledge graphs (KGs). Unlike passive diffusion, NaviRAG employs an agent that actively traverses the graph to act as a search-space pruning engine, identifying logical multi-hop reasoning paths. Technically, we introduce three key components: (1) Structure-Aware Query Expansion, which bridges the modality gap between unstructured queries and structured graph seeds for precise initialization; (2) Target-Driven Reward Shaping, which provides dense supervision based on semantic progress toward gold documents, effectively mitigating the sparse reward problem in large-scale graph traversal; and (3) a Multi-View Hybrid Reranking strategy that operates on the highly-pruned candidate subgraph, integrating policy confidence, semantic relevance, and global structural importance to ensure robust candidate selection. Extensive experiments on three multi-hop QA datasets and two single-hop QA datasets demonstrate that NaviRAG significantly outperforms baselines, achieving state-of-the-art performance in multi-hop QA while maintaining robustness in single-hop QA. Our code and data are available at https://github.com/CkingEW/NaviRAG.
Jinghong Lei, Wang Kun, Zhigang Chen et al.· Proceedings of the 32nd ACM...· 0 citations
Thanks to the development of basic models and the high quality of the data, the emergence of AI-generated content has
accelerated. Despite its incredible success, there are still challenges that are yet to be addressed in AI content generation, such as
the processing of long-trail information, maintaining up-to-date knowledge, addressing high inference and training expenses,
and addressing data leakage. The shift to address those challenges has been called Retrieval Augmented Generation (RAG).
RAG has brought the process of gathering information, which improves the process of data generation by recovering relevant
information from available data sources, resulting in robustness and accuracy. RAG has become the foundation of Natural
Language Processing (NLP) to effectively fill the gap between factual accuracy of knowledge and fluency of “Large Language
Models (LLMs)”. This study traces the root of RAG from its beginnings as a framework for knowledge-based tasks to its current
state as an agentic, modular and complex runtime of knowledge. This study is an in-depth analysis of the evolution of Naïve
RAG to Modular and Advanced RAG models, and the introduction of new innovations, such as self-reflection, dense vector
recovery, and the use of different models. They are then examined to provide detailed feedback on how to make RAG truly
dynamic and usable as a verifier when applying them to organizations.
Mallikarjunarao Sunke, S. Gudi, Sriharsha Gudi· International Journal for Re...· 0 citations
: The legal domain imposes unique demands on Large Language Models (LLMs), requiring high factual accuracy, precise statutory citation, and transparent reasoning. Retrieval-Augmented Generation (RAG) has emerged as a promising paradigm to mitigate hallucinations and improve knowledge grounding in legal applications. While general RAG systems have achieved success in other domains, legal scenarios present specific challenges due to the complexity of legal language, evolving jurisprudence, and jurisdictional variability. This survey provides a comprehensive overview of the technical evolution of legal RAG systems, categorizing existing approaches into four paradigms: Naive, Advanced, Modular, and Agentic RAG. Drawing on a broad set of recent legal Natural Language Processing (NLP) studies, we organize optimization strategies into three core stages of the RAG pipeline: Pre-retrieval, In-retrieval, and Post-retrieval. These stages highlight techniques adapted to legal use cases, such as document chunking, query rewriting, retrieval control, context filtering, and rationale-based response selection. By highlighting common architectural patterns and legal-specific adaptations, this survey aims to inform future research and practical deployment of robust, interpretable, and trustworthy legal LLMs.
Xin Li· Proceedings of the 3rd Inter...· 0 citations
Experimental results show that SAC-RAG reduces token consumption by 38%–58% at the cost of only a 1–2 percentage point EM drop, with EM actually improving after compression for reasoning-type questions, achieving the optimal quality–efficiency trade-off in terms of token consumption.
Deyu Zhang, Hongqiang Yu, Jinze Huo et al.· IEEE Access· 0 citations
Volume-based accuracy rewards retrieval-augmented generation (RAG) systems for guessing: a system that answers everything outscores one that declines when its knowledge base cannot support an answer. Building on the confidence-target analysis of Kalai et al. (2025), we present a penalty-aware evaluation framework for deployed RAG products, combining (i) asymmetric scoring (correct +1, wrong -4, abstain 0), (ii) knowledge-gap canaries, questions whose answers are verifiably absent from the knowledge base, so that any answer constitutes ungrounded generation from parametric memory, and (iii) a failure-attribution pipeline that separates retrieval, generation, and abstention-policy failures. Applying the framework to three commercial RAG systems and a no-retrieval baseline on SimpleQA-Verified (1,000 questions x 3 repeats, graded blind by a cross-family three-judge panel with 98.9% unanimity), we find that accuracy when answering is closely clustered across systems (97.0-98.0%), while canary violation rates differ roughly sixfold (16.7% vs. 98.1%). The systems are separated less by what they answer correctly than by whether they answer at all when they should not, and penalty-aware scoring reorders the volume-based ranking accordingly; the reordering is stable across penalty settings from k=1 to k=9. All code, configurations, transcripts, and judge votes are released for independent audit.
Alden Do Rosario, Hussein Younes, Felipe Pires· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
