DiscourseFlip is a graph-guided attack allocating limited poisoning budget to induce targeted opinion shifts over semantic query networks in black-box RAG.
super hub Canonical reference
Retrieval-Augmented Generation for Large Language Models: A Survey
Canonical reference. 86% of citing Pith papers cite this work as background.
402
Pith papers citing it
17
external citations · Pith
Background
86% of classified citations
abstract
Large Language Models (LLMs) showcase impressive capabilities but encounter challenges like hallucination, outdated knowledge, and non-transparent, untraceable reasoning processes. Retrieval-Augmented Generation (RAG) has emerged as a promising solution by incorporating knowledge from external databases. This enhances the accuracy and credibility of the generation, particularly for knowledge-intensive tasks, and allows for continuous knowledge updates and integration of domain-specific information. RAG synergistically merges LLMs' intrinsic knowledge with the vast, dynamic repositories of external databases. This comprehensive review paper offers a detailed examination of the progression of RAG paradigms, encompassing the Naive RAG, the Advanced RAG, and the Modular RAG. It meticulously scrutinizes the tripartite foundation of RAG frameworks, which includes the retrieval, the generation and the augmentation techniques. The paper highlights the state-of-the-art technologies embedded in each of these critical components, providing a profound understanding of the advancements in RAG systems. Furthermore, this paper introduces up-to-date evaluation framework and benchmark. At the end, this article delineates the challenges currently faced and points out prospective avenues for research and development.
hub tools
citation-role summary
background 63
method 5
dataset 2
citation-polarity summary
claims ledger
- abstract Large Language Models (LLMs) showcase impressive capabilities but encounter challenges like hallucination, outdated knowledge, and non-transparent, untraceable reasoning processes. Retrieval-Augmented Generation (RAG) has emerged as a promising solution by incorporating knowledge from external databases. This enhances the accuracy and credibility of the generation, particularly for knowledge-intensive tasks, and allows for continuous knowledge updates and integration of domain-specific information. RAG synergistically merges LLMs' intrinsic knowledge with the vast, dynamic repositories of exte
authors
co-cited works
representative citing papers
M³Att poisons medical multimodal RAG by pairing covert textual misinformation with query-agnostic visual perturbations that increase retrieval of the bad content, causing LLMs to generate clinically plausible but incorrect responses.
The first SoK on LLM-based AutoPT frameworks provides a six-dimension taxonomy of agent designs and a unified empirical benchmark evaluating 15 frameworks via over 10 billion tokens and 1,500 manually reviewed logs.
AgentClinic is a multimodal agent benchmark demonstrating that LLM diagnostic accuracy on MedQA drops to below one-tenth in sequential clinical simulations, with Claude-3.5 leading and large tool-use differences across models.
Manual context attachment produces a combinatorial collapse in AI task-success probability as personal corpus size and multi-document conjunctivity grow, while dynamic retrieval is insulated from that collapse, creating an inequality dimension complementary to agentic access.
Augmenting LLM search judges with historical QRI cards improves Spearman correlation with user preferences by ~5% overall (91% relative on disagreements) and 15% in multilingual settings, with better alignment to live A/B test outcomes.
DCCD combines document and token confidence signals to select and contrast positive/negative document streams, outperforming baselines on the new DRQA benchmark and standard multi-document QA tasks.
SOLAR is a learning-augmented policy for semantic cache replacement that achieves constant competitive ratio 3 and 5-75% gains over FIFO on retrieval workloads.
COHORT automates mitigation generation for network attacks via collaborative LLMs on emulated topologies with offensive replay evaluation, reporting 46.7% success rate that is 4.4 times higher than a single-agent baseline.
A fixed-iteration spreading activation with per-step cosine similarity gating enables query-aware KG retrieval as one database query, matching QAFD-RAG on MuSiQue while cutting latency.
ATHENA-R1 is an RL-trained agent using 212 biomedical tools that achieves 94.7% accuracy on drug reasoning and 82.9% on treatment reasoning tasks, outperforming GPT-5 by 17.8 and 10.7 points respectively.
LibEvoBench benchmark shows LLMs are version-oblivious on evolving APIs, with documentation helping but version specification not.
HERALD overlaps CPU-side sparse KV-cache selection with GPU-side block-denoising to hide offloading latency, achieving up to 2.28–2.47x decode throughput over dense GPU serving at 5–10% KV budgets.
User facts are internalized as surgical local edits to a hash-keyed Engram memory table with reasoning skill held in a shared adapter, claimed to match LoRA recall, improve indirect reasoning 5.6x on average, and compose across users with 33,000x smaller footprint than per-user adapters.
MetaSyn is a stage-level benchmark of 442 meta-analyses showing LLM agents retrieve up to 90.9% of eligible studies but include at most 52.7% in their final reports.
TAA-k finds query-adaptive retrieval cutoffs by first using knee detection to isolate a candidate window around the relevance-to-noise transition, then applying EVT goodness-of-fit tests inside that window.
Activation patching reveals that citation decisions in Llama-3.1-8B RAG are implemented by a distributed attributional ensemble of heads and layers; targeted interventions fix most missed and spurious citations on PopQA.
DCPM reorganizes LLM agent memory into a cognitive hierarchy driven by a synchronous daytime belief writer and an asynchronous nighttime schema engine, reporting gains on cross-session inference benchmarks.
LATTEArena introduces a 6-dimensional taxonomy and modular evaluation framework for LLM-powered tabular feature engineering, benchmarking 24 configurations to produce 17 findings on cost-effectiveness trade-offs with public artifacts.
QCFuse achieves full-prefill quality in RAG with 1.7x average prefill speedup over full prefill and 1.5x over ProphetKV via compressed query-aware cache fusion.
An empirical taxonomy of 11 top-level categories and 27 subcategories of runtime faults in MCP servers, derived via open coding of GitHub threads and validated by a survey of 55 developers.
Introduces CHARM framework that detects cascading hallucinations in agentic RAG at 89.4% rate with 5.3% false positives and reduces error propagation by 82.1% on multi-hop QA benchmarks.
Re-ranking retrieval candidates via a cross-encoder trained on continuous perturbation-based attribution scores improves citation faithfulness and gold-answer alignment in legal QA over semantic similarity.
SuperMemory-VQA provides 4,853 human-verified QA pairs from 52.9 hours of egocentric AI glasses recordings to benchmark AI systems on realistic long-horizon memory tasks including an unanswerable option.
citing papers explorer
-
DiscourseFlip: An Oblique Discourse-Level Opinion Manipulation Attack against Black-box Retrieval-Augmented Generation
DiscourseFlip is a graph-guided attack allocating limited poisoning budget to induce targeted opinion shifts over semantic query networks in black-box RAG.
-
Knowledge Poisoning Attacks on Medical Multi-Modal Retrieval-Augmented Generation
M³Att poisons medical multimodal RAG by pairing covert textual misinformation with query-agnostic visual perturbations that increase retrieval of the bad content, causing LLMs to generate clinically plausible but incorrect responses.
-
Hackers or Hallucinators? A Comprehensive Analysis of LLM-Based Automated Penetration Testing
The first SoK on LLM-based AutoPT frameworks provides a six-dimension taxonomy of agent designs and a unified empirical benchmark evaluating 15 frameworks via over 10 billion tokens and 1,500 manually reviewed logs.
-
AgentClinic: a multimodal agent benchmark to evaluate AI in simulated clinical environments
AgentClinic is a multimodal agent benchmark demonstrating that LLM diagnostic accuracy on MedQA drops to below one-tenth in sequential clinical simulations, with Claude-3.5 leading and large tool-use differences across models.
-
The Context Access Divide: Interaction-Level Architecture as a Complementary Dimension of Agentic Inequality
Manual context attachment produces a combinatorial collapse in AI task-success probability as personal corpus size and multi-document conjunctivity grow, while dynamic retrieval is insulated from that collapse, creating an inequality dimension complementary to agentic access.
-
As It Was: Aligning LLM Search Evaluation with Historical User Preferences
Augmenting LLM search judges with historical QRI cards improves Spearman correlation with user preferences by ~5% overall (91% relative on disagreements) and 15% in multilingual settings, with better alignment to live A/B test outcomes.
-
Dual-Confidence Contrastive Decoding for Retrieval-Augmented Generation
DCCD combines document and token confidence signals to select and contrast positive/negative document streams, outperforming baselines on the new DRQA benchmark and standard multi-document QA tasks.
-
When Classic Cache Policies Fail: Learning-Augmented Replacement for Semantic Retrieval Buffers
SOLAR is a learning-augmented policy for semantic cache replacement that achieves constant competitive ratio 3 and 5-75% gains over FIFO on retrieval workloads.
-
COHORT: Collaborative Orchestration for Hardening via Offensive Replay on Emulated Topologies
COHORT automates mitigation generation for network attacks via collaborative LLMs on emulated topologies with offensive replay evaluation, reporting 46.7% success rate that is 4.4 times higher than a single-agent baseline.
-
Query-Aware Spreading Activation for Multi-Hop Retrieval over Knowledge Graphs
A fixed-iteration spreading activation with per-step cosine similarity gating enables query-aware KG retrieval as one database query, matching QAFD-RAG on MuSiQue while cutting latency.
-
An AI agent for treatment reasoning over a biomedical tool universe
ATHENA-R1 is an RL-trained agent using 212 biomedical tools that achieves 94.7% accuracy on drug reasoning and 82.9% on treatment reasoning tasks, outperforming GPT-5 by 17.8 and 10.7 points respectively.
-
LibEvoBench: Probing Temporal Knowledge Stratification in Code Generation Models
LibEvoBench benchmark shows LLMs are version-oblivious on evolving APIs, with documentation helping but version specification not.
-
HERALD: High-Throughput Block Diffusion LLM Serving via CPU-GPU Cooperative KV Cache Retrieval
HERALD overlaps CPU-side sparse KV-cache selection with GPU-side block-denoising to hide offloading latency, achieving up to 2.28–2.47x decode throughput over dense GPU serving at 5–10% KV budgets.
-
User as Engram: Internalizing Per-User Memory as Local Parametric Edits
User facts are internalized as surgical local edits to a hash-keyed Engram memory table with reasoning skill held in a shared adapter, claimed to match LoRA recall, improve indirect reasoning 5.6x on average, and compose across users with 33,000x smaller footprint than per-user adapters.
-
MetaSyn: A Benchmark for LLM Agents on Meta-Analysis Articles from Nature Portfolio
MetaSyn is a stage-level benchmark of 442 meta-analyses showing LLM agents retrieve up to 90.9% of eligible studies but include at most 52.7% in their final reports.
-
Tail-Aware Adaptive-k: Query-Adaptive Context Selection for Retrieval-Augmented Generation
TAA-k finds query-adaptive retrieval cutoffs by first using knee detection to isolate a candidate window around the relevance-to-noise transition, then applying EVT goodness-of-fit tests inside that window.
-
How Do LLMs Cite? A Mechanistic Interpretation of Attribution in Retrieval-Augmented Generation
Activation patching reveals that citation decisions in Llama-3.1-8B RAG are implemented by a distributed attributional ensemble of heads and layers; targeted interventions fix most missed and spurious citations on PopQA.
-
Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents
DCPM reorganizes LLM agent memory into a cognitive hierarchy driven by a synchronous daytime belief writer and an asynchronous nighttime schema engine, reporting gains on cross-session inference benchmarks.
-
LATTEArena: An Evaluation Framework for LLM-powered Tabular Feature Engineering (Extended Version)
LATTEArena introduces a 6-dimensional taxonomy and modular evaluation framework for LLM-powered tabular feature engineering, benchmarking 24 configurations to produce 17 findings on cost-effectiveness trade-offs with public artifacts.
-
QCFuse: Query-Aware Cache Fusion via Compressed View for Efficient RAG Serving
QCFuse achieves full-prefill quality in RAG with 1.7x average prefill speedup over full prefill and 1.5x over ProphetKV via compressed query-aware cache fusion.
-
A Taxonomy of Runtime Faults in Model Context Protocol Servers
An empirical taxonomy of 11 top-level categories and 27 subcategories of runtime faults in MCP servers, derived via open coding of GitHub threads and validated by a survey of 55 developers.
-
Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation
Introduces CHARM framework that detects cascading hallucinations in agentic RAG at 89.4% rate with 5.3% false positives and reduces error propagation by 82.1% on multi-hop QA benchmarks.
-
Re-Ranking Through an Attribution Lens for Citation Quality in Legal QA
Re-ranking retrieval candidates via a cross-encoder trained on continuous perturbation-based attribution scores improves citation faithfulness and gold-answer alignment in legal QA over semantic similarity.
-
SuperMemory-VQA: An Egocentric Visual Question-Answering Benchmark for Long-Horizon Memory
SuperMemory-VQA provides 4,853 human-verified QA pairs from 52.9 hours of egocentric AI glasses recordings to benchmark AI systems on realistic long-horizon memory tasks including an unanswerable option.
-
Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets
Tool schema compression by 44-50% enables agentic RAG at 8K context where uncompressed schemas fail, with +20.5 pp exact match lift across models and scaling to over 800 tools.
-
Modernizing User Privacy Preference Measurement through GPPI: A GDPR-aligned Privacy Preference Item Bank
A 527-item GDPR-aligned privacy preference item bank was developed by extracting 669 statements from 99 GDPR articles and validating them through multi-round expert consensus and semantic clustering.
-
EngiAI: A Multi-Agent Framework and Benchmark Suite for LLM-Driven Engineering Design
EngiAI introduces a LangGraph-based multi-agent framework and a three-part benchmark suite for LLM-driven engineering design, reporting high task completion rates for proprietary models on Beams2D and Photonics2D problems.
-
UniPPTBench: A Unified Benchmark for Presentation Generation Across Diverse Input Settings
The paper presents UniPPTBench and UniPPTEval, a unified benchmark and scenario-aware evaluation framework for presentation generation from vague prompts, long documents, multimodal documents, and multi-source inputs.
-
Utility-Oriented Visual Evidence Selection for Multimodal Retrieval-Augmented Generation
Evidence utility is defined as information gain on the model's output distribution, with ranking by gain on a latent helpfulness variable shown equivalent to answer-space utility under mild assumptions, enabling a training-free surrogate framework that outperforms baselines.
-
Retrieval is Cheap, Show Me the Code: Executable Multi-Hop Reasoning for Retrieval-Augmented Generation
PyRAG turns multi-hop reasoning into executable Python code over retrieval tools for explicit, verifiable step-by-step RAG.
-
DeepRefine: Agent-Compiled Knowledge Refinement via Reinforcement Learning
DeepRefine refines agent-compiled knowledge bases via multi-turn abductive diagnosis and RL training with a GBD reward, yielding consistent downstream task gains.
-
EquiMem: Calibrating Shared Memory in Multi-Agent Debate via Game-Theoretic Equilibrium
EquiMem calibrates shared memory in multi-agent debate by computing a game-theoretic equilibrium from agent queries and paths, outperforming heuristics and LLM validators across benchmarks while remaining robust to adversarial agents.
-
SEM-RAG: Structure-Preserving Multimodal Graph Compilation and Entropy-Guided Retrieval for Telecommunication Standards
SEM-RAG compiles telecommunication standards into structure-preserving graphs and uses entropy-guided retrieval to reach 94.1% accuracy on TeleQnA and 93.8% on ORAN-Bench-13K while reducing indexing token usage compared to standard GraphRAG.
-
MANTRA: Synthesizing SMT-Validated Compliance Benchmarks for Tool-Using LLM Agents
MANTRA automatically synthesizes SMT-validated compliance benchmarks for LLM agents from natural language manuals and tool schemas, producing 285 tasks across 6 domains with minimal human effort.
-
LatentRAG: Latent Reasoning and Retrieval for Efficient Agentic RAG
LatentRAG performs agentic RAG by generating latent tokens for thoughts and subqueries in one forward pass, matching explicit methods' accuracy on seven benchmarks while reducing latency by ~90%.
-
Stateful Agent Backdoor
A stateful backdoor for LLM agents, modeled as a Mealy machine with a decomposition framework, enables incremental malicious actions across sessions and achieves 80-95% attack success rate on four models.
-
Privacy Without Losing Place: A Paradigm for Private Retrieval in Spatial RAGs
PAS encodes locations via relative anchors and bins to deliver roughly 370-400m adversarial error in spatial RAG while retaining over half the baseline retrieval performance and keeping generation quality robust.
-
Telegraph English: Semantic Prompt Compression via Structured Symbolic Rewriting
Telegraph English compresses prompts via structured symbolic rewriting into atomic facts, achieving roughly 50% token reduction with 99.1% key-fact accuracy on LongBench-v2 and outperforming token-deletion baselines across models.
-
RAG over Thinking Traces Can Improve Reasoning Tasks
Retrieving structured thinking traces as a corpus improves reasoning performance on AIME, LiveCodeBench, and GPQA over standard RAG or no retrieval.
-
Chain of Evidence: Pixel-Level Visual Attribution for Iterative Retrieval-Augmented Generation
Chain of Evidence introduces a retriever-agnostic visual attribution method for iRAG that reasons over document screenshots with VLMs to output precise bounding boxes, outperforming text baselines on Wiki-CoE and SlideVQA.
-
E-MIA: Exam-Style Black-Box Membership Inference Attacks against RAG Systems
E-MIA converts document details into four types of exam questions and aggregates the RAG's answers into a membership score that separates member and non-member documents better than prior similarity-based or probe-based attacks.
-
ReLay: Personalized LLM-Generated Plain-Language Summaries for Better Understanding, but at What Cost?
Personalized LLM-generated plain language summaries improve lay readers' comprehension and quality ratings but increase risks of reinforcing biases and introducing hallucinations compared to static expert summaries.
-
TADI: Tool-Augmented Drilling Intelligence via Agentic LLM Orchestration over Heterogeneous Wellsite Data
TADI shows that domain-specialized tools orchestrated by an LLM over dual structured and semantic databases can convert heterogeneous wellsite data into evidence-grounded drilling intelligence, with tool design mattering more than model scale.
-
Quantifying Prior Dominance in RAG Systems
Introduces NCU metric using token log-probabilities and finds small language models match or outperform larger ones in strict factual RAG extraction, while commercial APIs show high prior dominance and negative transfer.
-
When to Retrieve During Reasoning: Adaptive Retrieval for Large Reasoning Models
ReaLM-Retrieve uses step-level uncertainty to trigger retrievals during reasoning, achieving 10.1% better F1 scores and 47% fewer calls on multi-hop QA benchmarks.
-
RepoDoc: A Knowledge Graph-Based Framework to Automatic Documentation Generation and Incremental Updates
RepoDoc uses a repository knowledge graph with module clustering and semantic impact propagation to generate more complete documentation 3x faster with 85% fewer tokens and handle incremental updates 73% faster than prior LLM-based tools.
-
Context-Augmented Code Generation: How Product Context Improves AI Coding Agent Decision Compliance by 49%
Adding product context retrieval to AI coding agents raises decision compliance from 46% to 95% on a new benchmark of 8 tasks with 41 weighted decision points.
-
XGRAG: A Graph-Native Framework for Explaining KG-based Retrieval-Augmented Generation
XGRAG uses graph perturbations to quantify component contributions in GraphRAG and achieves 14.81% better explanation quality than text-based baselines on QA datasets, with correlations to graph centrality.
-
Similar Users-Augmented Interest Network
SUIN improves CTR prediction by augmenting target user sequences with similar users' behaviors via embedding-based retrieval, user-specific position encoding, and user-aware target attention.
-
Uncertainty Propagation in LLM-Based Systems
This paper introduces a systems-level conceptual framing and a three-level taxonomy (intra-model, system-level, socio-technical) for uncertainty propagation in compound LLM applications, along with engineering insights and open challenges.