MemCoE learns memory organization guidelines via contrastive feedback and then trains a guideline-aligned RL policy for memory updates, yielding consistent gains on personalization benchmarks.
hub Canonical reference
A survey of graph retrieval-augmented generation for customized large language models
Canonical reference. 88% of citing Pith papers cite this work as background.
30
Pith papers citing it
Background
88% of classified citations
hub tools
citation-role summary
background 8
citation-polarity summary
roles
background 8representative citing papers
Repurposing competency questions as runtime executable plans creates a controlled neuro-symbolic RAG architecture that produces evidence-closed stories from knowledge graphs.
AtomicRAG replaces chunk-based and triple-based GraphRAG with atom-entity graphs that store facts as atomic units and use personalized PageRank plus relevance filtering to achieve higher retrieval accuracy and reasoning robustness on five benchmarks.
M³KG-RAG improves multimodal reasoning in large language models by constructing multi-hop knowledge graphs and selectively pruning retrieved context with GRASP.
SHACR is a graph-augmented framework that grounds LLMs in a formal knowledge graph to unify logical, semantic, and physical conflict detection in IoT automation, raising F1 from 0.59 to 0.95 on a 203-rule testbed.
MemGraphRAG uses a memory-based multi-agent system for globally consistent graph construction from fragmented corpora plus a memory-aware hierarchical retriever, claiming better benchmark performance than prior GraphRAG methods at similar cost.
MoG uses hub graphs for shared context and sparsely activates expert graphs with a topology-aware router, reporting over 20% relative gains on MuSiQue.
TERGAD augments graph anomaly detection by converting node topological properties into LLM-generated semantic embeddings that are fused with original attributes via a gated dual-branch autoencoder for joint reconstruction-based anomaly scoring.
TeleCom-Bench reveals LLMs reach 90% on telecom intent and entity tasks but drop to 30% on solution generation and root cause analysis in live network scenarios.
SAGE is a self-evolving agentic graph-memory engine that dynamically constructs and refines structured memory graphs via writer-reader feedback, yielding performance gains on multi-hop QA, open-domain retrieval, and long-term agent benchmarks.
LARAG improves RAG answer quality on hyperlinked technical documentation by using author-defined links for retrieval, achieving higher BERTScore while using fewer chunks and tokens than standard embedding-based RAG.
RAGP models prompt compression as redundancy-aware pruning on a multiplex graph using Lévy walks, achieving 49.3 average on LongBench at 4x compression versus 48.8 for LongLLMLingua at 3x.
EvoRAG adds a feedback-driven backpropagation step that attributes response quality to individual knowledge-graph triplets and updates the graph to raise reasoning accuracy by 7.34 percent over prior KG-RAG methods.
CS-RAG is a GraphRAG framework that plans queries as ordered atomic constraints, uses anchor-relation aware retrieval, applies sufficiency checks, and falls back to text recovery to reduce drift and hallucination from imperfect KGs.
An iterative feedback-driven GraphRAG architecture produces higher semantic quality and relevance on HotPotQA queries than single-shot baselines.
KaSLA applies knapsack optimization hierarchically to schema linking for LLM text-to-SQL, claiming better results than large models and improved SQL generation on Spider and BIRD.
CS-PQ optimizes PQ construction on CPUs via vectorized SIMD across centroids and cache-restructured pipelines, claiming 10.7x speedup over prior CPU methods with no accuracy loss on large datasets.
A survey that organizes existing work on LLM-based agents around code as the central harness, structured in three layers of interfaces, mechanisms, and multi-agent scaling, with applications across domains and listed open challenges.
Active information seeking via search tools, when combined with multi-candidate context pruning during training, produces consistent gains on translation, health, and reasoning tasks over naive tool addition or no-tool baselines.
AVA is a specialized GenAI platform for development policy research that provides verifiable syntheses from World Bank reports and is associated with 2.4-3.9 hours of weekly time savings in a large-scale user evaluation.
STAR is a semantic-tuned and tail-adaptive retriever for GraphRAG that uses cross-attention interaction learning and path-weighted contrastive learning to mitigate Semantic Shortcut Bias and Long-Tail Path Bias, reporting 1.8% retrieval and 2.2% QA gains.
MemOS introduces a unified memory management framework for LLMs using MemCubes to handle and evolve different memory types for improved controllability and evolvability.
FeLoG achieves 27.9x average speedup and over 53% communication reduction in distributed graph embedding by using feedback-coupled sampling, activity-aware communication, and round-interleaved pipelining.
Didact is a RAG-based prototype with an Evidence Rail for conversational capability discovery from integrated Australian defence documents and research publications.
citing papers explorer
-
Learning How and What to Memorize: Cognition-Inspired Two-Stage Optimization for Evolving Memory
MemCoE learns memory organization guidelines via contrastive feedback and then trains a guideline-aligned RL policy for memory updates, yielding consistent gains on personalization benchmarks.
-
Competency Questions as Executable Plans: a Controlled RAG Architecture for Cultural Heritage Storytelling
Repurposing competency questions as runtime executable plans creates a controlled neuro-symbolic RAG architecture that produces evidence-closed stories from knowledge graphs.
-
AtomicRAG: Atom-Entity Graphs for Retrieval-Augmented Generation
AtomicRAG replaces chunk-based and triple-based GraphRAG with atom-entity graphs that store facts as atomic units and use personalized PageRank plus relevance filtering to achieve higher retrieval accuracy and reasoning robustness on five benchmarks.
-
M$^3$KG-RAG: Multi-hop Multimodal Knowledge Graph-enhanced Retrieval-Augmented Generation
M³KG-RAG improves multimodal reasoning in large language models by constructing multi-hop knowledge graphs and selectively pruning retrieved context with GRASP.
-
SHACR: A Graph-Augmented Semi-Autonomous Framework for Multi-Class Conflict Resolution in Smart Home IoT Automation
SHACR is a graph-augmented framework that grounds LLMs in a formal knowledge graph to unify logical, semantic, and physical conflict detection in IoT automation, raising F1 from 0.59 to 0.95 on a 203-rule testbed.
-
MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation
MemGraphRAG uses a memory-based multi-agent system for globally consistent graph construction from fragmented corpora plus a memory-aware hierarchical retriever, claiming better benchmark performance than prior GraphRAG methods at similar cost.
-
MoG: Mixture of Experts for Graph-based Retrieval-Augmented Generation
MoG uses hub graphs for shared context and sparsely activates expert graphs with a topology-aware router, reporting over 20% relative gains on MuSiQue.
-
TERGAD: Structure-Aware Text-Enhanced Representations for Graph Anomaly Detection
TERGAD augments graph anomaly detection by converting node topological properties into LLM-generated semantic embeddings that are fused with original attributes via a gated dual-branch autoencoder for joint reconstruction-based anomaly scoring.
-
TeleCom-Bench: How Far Are Large Language Models from Industrial Telecommunication Applications?
TeleCom-Bench reveals LLMs reach 90% on telecom intent and entity tasks but drop to 30% on solution generation and root cause analysis in live network scenarios.
-
SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory
SAGE is a self-evolving agentic graph-memory engine that dynamically constructs and refines structured memory graphs via writer-reader feedback, yielding performance gains on multi-hop QA, open-domain retrieval, and long-term agent benchmarks.
-
LARAG: Link-Aware Retrieval Strategy for RAG Systems in Hyperlinked Technical Documentation
LARAG improves RAG answer quality on hyperlinked technical documentation by using author-defined links for retrieval, achieving higher BERTScore while using fewer chunks and tokens than standard embedding-based RAG.
-
Mapping Text to Multiplex Graph: Prompt Compression as L\'evy Walk-Guided Graph Pruning
RAGP models prompt compression as redundancy-aware pruning on a multiplex graph using Lévy walks, achieving 49.3 average on LongBench at 4x compression versus 48.8 for LongLLMLingua at 3x.
-
EvoRAG: Making Knowledge Graph-based RAG Automatically Evolve through Feedback-driven Backpropagation
EvoRAG adds a feedback-driven backpropagation step that attributes response quality to individual knowledge-graph triplets and updates the graph to raise reasoning accuracy by 7.34 percent over prior KG-RAG methods.
-
Toward Robust GraphRAG: Mitigating Retrieval Drift and Hallucination from Imperfect Knowledge Graphs
CS-RAG is a GraphRAG framework that plans queries as ordered atomic constraints, uses anchor-relation aware retrieval, applies sufficiency checks, and falls back to text recovery to reduce drift and hallucination from imperfect KGs.
-
KGiRAG: An Iterative GraphRAG Approach for Responding Sensemaking Queries
An iterative feedback-driven GraphRAG architecture produces higher semantic quality and relevance on HotPotQA queries than single-shot baselines.
-
Knapsack Optimization-based Schema Linking for LLM-based Text-to-SQL Generation
KaSLA applies knapsack optimization hierarchically to schema linking for LLM text-to-SQL, claiming better results than large models and improved SQL generation on Spider and BIRD.
-
CS-PQ: Cache-Friendly SIMD Product Quantization for Large-Scale ANNS Index Construction
CS-PQ optimizes PQ construction on CPUs via vectorized SIMD across centroids and cache-restructured pipelines, claiming 10.7x speedup over prior CPU methods with no accuracy loss on large datasets.
-
Code as Agent Harness
A survey that organizes existing work on LLM-based agents around code as the central harness, structured in three layers of interfaces, mechanisms, and multi-agent scaling, with applications across domains and listed open challenges.
-
Context Training with Active Information Seeking
Active information seeking via search tools, when combined with multi-candidate context pruning during training, produces consistent gains on translation, health, and reasoning tasks over naive tool addition or no-tool baselines.
-
Learning from AVA: Early Lessons from a Curated and Trustworthy Generative AI for Policy and Development Research
AVA is a specialized GenAI platform for development policy research that provides verifiable syntheses from World Bank reports and is associated with 2.4-3.9 hours of weekly time savings in a large-scale user evaluation.
-
STAR: Semantic-Tuned and Tail-Adaptive Retriever for Graph-Augmented Generation
STAR is a semantic-tuned and tail-adaptive retriever for GraphRAG that uses cross-attention interaction learning and path-weighted contrastive learning to mitigate Semantic Shortcut Bias and Long-Tail Path Bias, reporting 1.8% retrieval and 2.2% QA gains.
-
MemOS: A Memory OS for AI System
MemOS introduces a unified memory management framework for LLMs using MemCubes to handle and evolve different memory types for improved controllability and evolvability.
-
FeLoG: Scalable and Efficient Distributed Graph Embedding with Feedback Loop Mechanism
FeLoG achieves 27.9x average speedup and over 53% communication reduction in distributed graph embedding by using feedback-coupled sampling, activity-aware communication, and round-interleaved pipelining.
-
Didact: A Cross-Domain Capability Discovery System for Defence
Didact is a RAG-based prototype with an Evidence Rail for conversational capability discovery from integrated Australian defence documents and research publications.
-
A Community Survey on SHACL and ShEx: Briding Gaps in RDF Validation
A survey of RDF validation users finds the technology is valued for data quality but needs better documentation, tool support, performance, and expressiveness for complex tasks.
-
Accurate, Efficient, and Explainable Deep Learning Approaches for Environmental Science Problems
The work introduces WaLeF/FIDLAr for flood forecasting, CoDiCast for probabilistic weather, and Hypercube-RAG for explainable environmental QA, claiming superior accuracy, efficiency, and interpretability over baselines.
-
Behavioral Intelligence Platforms: From Event Streams to Autonomous Insight via Probabilistic Journey Graphs, Behavioral Knowledge Extraction, and Grounded Language Generation
BIP turns event streams into autonomous insights by modeling journeys as absorbing Markov chains, extracting facts via knowledge graphs, and generating narratives constrained to verified data.
-
CogniVerse: Revolutionizing Multi-Modal Retrieval-Augmented Generation with Cognitive Reflection and Geometric Reasoning
CogniVerse is a proposed MMRAG framework that combines cognitive reflection for retrieval filtering, Riemannian manifold alignment plus spectral graphs for retrieval, and optimal transport loss for generation, claiming better accuracy, coherence, and lower latency than prior systems.
-
The Dynamic Gist-Based Memory Model (DGMM): A Memory-Centric Architecture for Artificial Intelligence
DGMM is proposed as an explicit graph-structured memory architecture for AI that enables persistent episodic memory, cue-based recall, and context-dependent interpretation without retraining.
- Retrieval-Augmented Generation Must Move Beyond Factual Grounding to Represent Diverse Opinions