Shadowgraph
TrendingAbout
Sign in

ngl567/KGR-Survey

A Survey of Task-Oriented Knowledge Graph Reasoning: Status, Applications, and Prospects

81

34 commits

updated Mar 11, 2026

arxiv.org/abs/2506.11012

See the code

README

KGR-Survey

Awesome License: MIT preprint

A Survey of Task-Oriented Knowledge Graph Reasoning: Status, Applications, and Prospects Paper

🚀Main contribution of this survey: This survey provides a more comprehensive perspective on the research of KGR by categorizing approaches based on primary reasoning tasks, downstream application tasks, and potential challenging reasoning tasks. Besides, we explore advanced techniques, such as large language models (LLMs), and their impact on KGR. This work aims to highlight key research trends and outline promising future directions in the field of KGR.

🙌Key characteristics of this repository: Unlike other outstanding review repositories of the knowledge graph reasoning field, we not only provide a comprehensive review but also strive to offer the official publication abstract page for each paper. This includes not only the official publication version of the paper but also additional resources such as author information, videos, datasets, supplementary materials, and BibTeX citations.

If this repository is useful for you, please kindly cite the corresponding survey paper:

@misc{niu2025kgrsurvey,
  author       = {Guanglin Niu and Bo Li and Yangguang Lin},
  title        = {A Survey of Task‐Oriented Knowledge Graph Reasoning: Status, Applications, and Prospects},
  year         = {2025},
  eprint       = {arXiv:2506.11012},
  archivePrefix= {arXiv},
  primaryClass = {cs.AI},
  url          = {https://arxiv.org/abs/2506.11012}
}

The comprehensive overview framework of our survey is presented as following. The same number (①-⑨) indicates that different approaches share similar ideas, and the keywords corresponding to each number are provided at the bottom of the figure.

The illustration of the six primary KGR tasks

Illustration of primary KGR tasks

🔥 News

Content

Survey Papers

TitleConference/JournalYearCharacteristicPaper
A survey of task-oriented knowledge graph reasoning: status, applications, and prospectsarXiv2025Task-oriented KGRlink
Knowledge graph embedding: a survey from the perspective of representation spacesACM Computer Survey2024Embedding Spaceslink
A survey of knowledge graph reasoning on graph types: Static, dynamic, and multi-modalIEEE TPAMI2024Graph Typeslink
Negative sampling in knowledge graph representation learning: a reviewarXiv2024Negative Samplinglink
Overview of knowledge reasoning for knowledge graphNeurocomputering2024Causal Reasoninglink
A survey on temporal knowledge graph: representation learning and applicationsarXiv2024Temporal Reasoninglink
A survey on temporal knowledge graph completion: taxonomy, progress, and prospectsarXiv2023Temporal Reasoninglink
Generalizing to unseen elements: a survey on knowledge extrapolation for knowledge graphsIJCAI2023Unseen Elementslink
A survey on few-shot knowledge graph completion with structural and commonsense knowledgearXiv2023Commonsenselink
Beyond transduction: a survey on inductive, few shot, and zero shot link prediction in knowledge graphsarXiv2023Few-shot & Inductivelink
A comprehensive overview of knowledge graph completionKnowledge-Based System2022Multi-modal & Hyper-relationlink
Knowledgegraph reasoning with logics and embeddings: survey and perspectivearXiv2022Logics and Embeddingslink

⬆️

Static Single-Step KGR

KGE-based KGR Model

The illustration of five representative KGE models

Illustration of primary KGR tasks

Translation or Tensor Decomposition-Based KGE Models

ModelTitleConference/JournalYearPaper
TransETranslating embeddings for modeling multi-relational dataNIPS2013link
TransHKnowledge graph embedding by translating on hyperplanesAAAI2014link
TransRLearning entity and relation embeddings for knowledge graph completionAAAI2015link
TransDKnowledge graph embedding via dynamic mapping matrixACL2015link
TranSparseKnowledge graph completion with adaptive sparse transfer matrixAAAI2016link
PairEPairRE: Knowledge graph embeddings via paired relation vectorsACL2021link
TransATransA: An adaptive approach for knowledge graph embeddingarXiv2015link
KG2ELearning to represent knowledge graphs with Gaussian embeddingCIKM2015link
ManifoldEFrom one point to a manifold: Knowledge graph embedding for precise link predictionIJCAI2016link
TorusETorusE: Knowledge graph embedding on a Lie groupAAAI2018link
PoincaréPoincare embeddings for learning hierarchical representationsNIPS2017link
MuRPMulti-relational Poincare graph embeddingsNIPS2019link
HAKELearning hierarchy-aware knowledge graph embeddings for link predictionAAAI2020link
H2EKnowledge graph representation via hierarchical hyperbolic neural graph embeddingIEEE Big Data2021link
HBEHyperbolic hierarchy-aware knowledge graph embedding for link predictionEMNLP2021link
RotatERotatE: Knowledge graph embedding by relational rotation in complex spaceICLR2019link
QuatEQuaternion knowledge graph embeddingNIPS2019link
DualEDual quaternion knowledge graph embeddingsAAAI2021link
RESCALA three-way model for collective learning on multi-relational dataICML2011link
PITF-BPRPredicting RDF triples in incomplete knowledge bases with tensor factorizationSAC2012link
DistMultEmbedding entities and relations for learning and inference in knowledge basesICLR2015link
ComplExComplex embeddings for simple link predictionICML2016link
HolEHolographic embeddings of knowledge graphsAAAI2016link

(Graph) Neural Network-based Models

ModelTitleConference/JournalYearPaper
NTNReasoning with neural tensor networks for knowledge base completionNIPS2013link
SMEA semantic matching energy function for learning with multi-relational dataMachine Learning2014link
NAMProbabilistic reasoning via deep learning: Neural association modelsarXiv2016link
ConvEConvolutional 2D knowledge graph embeddingsAAAI2018link
ConvKBA novel embedding model for knowledge base completion based on convolutional neural networkNAACL2018link
GNN SurveyA comprehensive survey on graph neural networksIEEE TNNLS2021link
R-GCNModeling relational data with graph convolutional networksESWC2018Link
SACNEnd-to-end structure-aware convolutional networks for knowledge base completionAAAI2019link
KBGATLearning attention-based embeddings for relation prediction in knowledge graphsACL2019link
KE-GCNKnowledge embedding based graph convolutional networkThe Web Conference2021link

Transformer-based Models

ModelTitleConference/JournalYearPaper
KG-BERTModeling relational data with graph convolutional networksESWC2018Link
R-MeNA relational memory-based embedding model for triple classification and search personalizationACL2021link
CoKECoKE: Contextualized knowledge graph embeddingarXiv2019link
HittERHittER: Hierarchical transformers for knowledge graph embeddingsEMNLP2021link
GenKGCFrom discrimination to generation: Knowledge graph completion with generative transformerWWW2022link
iHTPre-training transformers for knowledge graph completionarXiv2023link
SimKGCSimKGC: Simple contrastive knowledge graph completion with pre-trained language modelsACL2022link
StARStructure-augmented text representation learning for efficient knowledge graph completionWWW2021link
KoPAMaking large language models perform better in knowledge graph completionarXiv2023link
KICGPTKICGPT: Large language model with knowledge in context for knowledge graph completionEMNLP2023link
RelphormerRelphormer: Relational graph transformer for knowledge graph representationsNeurocomputing2024link
LGKGRLGKGR: A knowledge graph reasoning model using LLMs augmented GNNsNeurocomputing2025Link

⬆️

Ontology-Enhanced KGE Models

ModelTitleConference/JournalYearPaper
JOIEUniversal representation learning of knowledge bases by jointly embedding instances and ontological conceptsKDD2019Link
Nickel et al.Factorizing YAGO: Scalable machine learning for linked dataWWW2012link
CISSEmbedding two-view knowledge graphs with class inheritance and structural similarityKDD2024link
Wang et al.An ontology-enhanced knowledge graph embedding methodICCPR2024link
Concept2BoxConcept2Box: Joint geometric embeddings for learning two-view knowledge graphsACL2023link
CAKECAKE: A scalable commonsense-aware framework for multi-view knowledge graph completionACL2022link
SSESemantically smooth knowledge graph embeddingACL2015link
TKRLRepresentation learning of knowledge graphs with hierarchical typesIJCAI2016link
TransETTransET: Knowledge graph embedding with entity typesElectronics2021link
AutoETERAutoETER: Automated entity type representation for knowledge graph embeddingEMNLP2020link

Path-Enhanced KGE Models

ModelTitleConference/JournalYearPaper
Path-RNNCompositional vector space models for knowledge base completionACL2015link
PTransEModeling relation paths for representation learning of knowledge basesEMNLP2015Link
PRNA path-based relation networks model for knowledge graph completionExpert Systems with Applications2021link
OPTransERepresentation learning with ordered relation paths for knowledge graph completionEMNLP-IJCNLP2019link
TransE&RWModeling relation paths for knowledge base completion via joint adversarial trainingKnowledge Based Systems2020link
HARPAHARPA: hierarchical attention with relation paths for knowledge graph embedding adversarial learningData Mining and Knowledge Discovery2023link
RPJERule-guided compositional representation learning on knowledge graphsAAAI2020link
PARLAttention-aware path-based relation extraction for medical knowledge graphSmart Computing and Communication2017link
Das et al.Chains of reasoning over entities, relations, and text using recurrent neural networksEACL2017link
Jiang et al.Attentive path combination for knowledge graph completionMachine Learning Research2017link
CPConvKEA confidence-aware and path-enhanced convolutional neural network embedding framework on noisy knowledge graphNeurocomputing2023link
PaSKoGEPath-specific knowledge graph embeddingKnowledge-based Systems2018link
Jagvaral et al.Path-based reasoning approach for knowledge graph completion using CNN-BiLSTM with attention mechanismExpert Systems with Applications2020link
PathConRelational message passing for knowledge graph completionKDD2021link
PTrustEPTrustE: A high-accuracy knowledge graph noise detection method based on path trustworthiness and triple embeddingKnowledge-based Systems2022link
TAPRModeling relation paths for knowledge graph completionIEEE TKDE2021link
Niu et al.Joint semantics and data-driven path representation for knowledge graph reasoningNeurocomputing2022link

⬆️

Negative Sampling for KGE

The illustration of six types of negative sampling strategies

Illustration of negative sampling strategies

ModelTitleConference/JournalYearPaper
Local Closed-World AssumptionKnowledge Vault: A web scale approach to probabilistic knowledge fusionKDD2014link
NS SurveyNegative sampling in knowledge graph representation learning: A reviewarXiv2023link
Uniform SamplingKnowledge graph embedding by translating on hyperplanesAAAI2014link
KBGANKBGAN: Adversarial learning for knowledge graph embeddingsNAACL2018Link
Self-AdvRotatE: Knowledge graph embedding by relational rotation in complex spaceICLR2019link
Batch NSPytorch-BigGraph: A large scale graph embedding systemMachine Learning and Systems2019link
Bernoulli NSAn interpretable knowledge transfer model for knowledge base completionACL2017link
Zhang et al.A novel negative sample generating method for knowledge graph embeddingEWSN2019link
SparseNSGA novel negative sampling based on frequency of relational association entities for knowledge graph embeddingJournal of Web Engineering2021link
IGANIncorporating GAN for negative sampling in knowledge representation learningAAAI2018link
GraphGANGraphGAN: Graph representation learning with generative adversarial netsAAAI2018link
KSGANA knowledge selective adversarial network for link prediction in knowledge graphNLPCC2019link
RUGAImproving knowledge graph completion using soft rules and adversarial learningChinese Journal of Electronics2021link
LASAdversarial knowledge representation learning without external modelIEEE Access2019link
ASARelation-aware graph attention model with adaptive self-adversarial trainingAAAI2021link
ANKnowledge graph embedding based on adaptive negative samplingICPSEE2019link
EANSEntity aware negative sampling with auxiliary loss of false negative prediction for knowledge graph embeddingarXiv2022link
Truncated NSFusing attribute character embeddings with truncated negative sampling for entity alignmentElectronics2023link
DNSDistributional negative sampling for knowledge base completionarXiv2019link
ESNSEntity similarity-based negative sampling for knowledge graph embeddingPRICAI2022Link
RCWCKGBoost: A classification-based knowledge base completion method with negative samplingPattern Recognition Letters2022link
Conditional SamplingConditional constraints for knowledge graph embeddingsDL4KG2020link
LEMONLEMON: LanguagE MOdel for negative sampling of knowledge graph embeddingsarXiv preprint2022Link
NSCachingNSCaching: Simple and efficient negative sampling for knowledge graph embeddingICDE2019Link
MDNcachingMDNcaching: A strategy to generate quality negatives for knowledge graph embeddingIEA/AIE2022Link
Op-TransOp-Trans: An optimization framework for negative sampling and triplet-mapping properties in knowledge graph embeddingApplied Sciences2023Link
NS-KGEEfficient non-sampling knowledge graph embeddingThe Web Conference2021Link

⬆️

Open-Source Library for KGE

LibraryImplementationKey FeaturesGitHub Repository
OpenKEPytorch, TensorFlow, C++Efficiently implements fundamental operations such as data loading, negative sampling, and performance evaluation using C++ for high performance.https://github.com/thunlp/OpenKE
AmpliGraphTensorFlowProvides a Keras-style API with improved efficiency over OpenKE.https://github.com/Accenture/AmpliGraph
torchKGEPytorchAchieves twice the efficiency of OpenKE and five times that of AmpliGraph.https://github.com/torchkge-team/torchkge
LibKGEPytorchEnables direct configuration of hyperparameters and model settings via configuration files.https://github.com/uma-pi1/kge
KB2EC++One of the earliest KGE libraries and the predecessor of OpenKE.https://github.com/thunlp/KB2E
scikit-kgePythonImplements multiple classical KGE models and supports a novel negative sampling strategy.https://github.com/mnick/scikit-kge
NeuralKGPytorchIntegrates KGE techniques with graph neural networks (GNNs) and rule-based reasoning models.https://github.com/zjukg/NeuralKG
PyKEENPytorchOffers 37 datasets, 40 KGE models, 15 loss functions, 6 regularization mechanisms, and 3 negative sampling strategies.https://github.com/pykeen/pykeen
Pykg2vecPytorch, TensorFlowSupports automated hyperparameter tuning, exports KG embeddings in TSV or RDF formats, and provides visualization for performance evaluation.https://github.com/Sujit-O/pykg2vec
μKGPytorch, TensorFlowSupports multi-process execution and GPU-accelerated computation, making it well-suited for large-scale KGs.https://github.com/nju-websoft/muKG
DGL-KEPytorch, MXNetOptimized for execution on CPU and GPU clusters, offering high scalability for large-scale KGs.https://github.com/awslabs/dgl-ke
GraphVitePytorchProvides efficient large-scale embedding learning, supports visualization of graph data, and enables multi-processing and GPU parallelization.https://github.com/DeepGraphLearning/graphvite
PBGPytorchDesigned for distributed training, capable of handling KGs with billions of entities and trillions of edges.https://github.com/facebookresearch/PyTorch-BigGraph

⬆️

Logic Rule-based KGR Model

Rule Learning for KG

ModelTitleConference/JournalYearPaper
FOILLearning logical definitions from relationsMachine Learning1990link
MDIEInverse entailment and progolNew Generation Computing1995link
InspireBest-effort inductive logic programming via fine-grained cost-based hypothesis generationMachine Learning2018link
Neural-Num-LPDifferentiable learning of numerical rules in knowledge graphsICLR2020link
AMIE+Fast rule mining in ontological knowledge bases with AMIE+VLDB Journal2015link
ScaLeKBScaLeKB: Scalable learning and inference over large knowledge basesVLDB Journal2016link
RDF2rulesRDF2Rules: Learning rules from RDF knowledge bases by mining frequent predicate cyclesarXiv2015link
SWARMSWARM: An approach for mining semantic association rules from semantic web dataPRICAI2016link
RudikRudik: Rule discovery in knowledge basesPVLDB2018link
RuLESRule learning from knowledge graphs guided by embedding modelsESWC2018link
EvodaRule learning over knowledge graphs with genetic logic programmingICDE2022link
NeuralLPDifferentiable learning of logical rules for knowledge base reasoningNeurIPS2017link
DRUMDRUM: End-to-end differentiable rule mining on knowledge graphsNeurIPS2019link
RLvLRAn embedding-based approach to rule learning in knowledge graphsIEEE TKDE2019link
RNNLogicRNNLogic: learning logic rules for reasoning on knowledge graphsICLR2021link
RARLRelatedness and TBox-driven rule learning in large knowledge basesAAAI2020link
RuleformerRuleformer: context-aware rule mining over knowledge graphCOLING2022link
Ott et al.Rule-based knowledge graph completion with canonical modelsCIKM2023link

Neural-Symbolic KGR

ModelTitleConference/JournalYearPaper
KALEJointly embedding knowledge graphs and logical rulesEMNLP2016link
RUGEKnowledge graph embedding with iterative guidance from soft rulesAAAI2018link
RulERulE: Knowledge graph reasoning with rule embeddingFindings of ACL2024link
RPJERule-guided compositional representation learning on knowledge graphsAAAI2020link
IterEIteratively learning embeddings and rules for knowledge graph reasoningWWW2019link
UniKERUniKER: A unified framework for combining embedding and definite Horn rule reasoning for knowledge graph inferenceEMNLP2021link
EngineKGPerform like an engine: A closed-loop neural-symbolic learning framework for knowledge graph inferenceCOLING2022link
Taxonomy of static single-step KGR approaches

Single-step KGR

⬆️

Static Multi-Step KGR

Random Walk-based Model

ModelTitleConference/JournalYearPaper
PRARelational retrieval using a combination of path-constrained random walksMachine Learning2010link
Lao et al. 1Random walk inference and learning in a large scale knowledge baseEMNLP2011link
Lao et al. 2Reading the web with learned syntactic-semantic inference rulesEMNLP2012link
Gardner et al.Improving learning and inference in a large knowledge-base using latent syntactic cuesEMNLP2013link
CPRAKnowledge base completion via coupled path rankingACL2016link
C-PRContext-aware path ranking for knowledge base completionIJCAI2017link
A*NetA*Net: a scalable path-based reasoning approach for knowledge graphsNeurIPS2024link
SFEEfficient and expressive knowledge base completion using subgraph feature extractionEMNLP2015link
PathConRelational message passing for knowledge graph completionKDD2021link

Reinforcement Learning-based Model

ModelTitleConference/JournalYearPaper
DeepPathDeepPath: a reinforcement learning method for knowledge graph reasoningEMNLP2017link
MINERVAGo for a walk and arrive at the answer: Reasoning over paths in knowledge bases using reinforcement learningICLR2018link
DIVAVariational knowledge graph reasoningNAACL2018link
MultiHopKGMulti-hop knowledge graph reasoning with reward shapingEMNLP2018link
M-WalkM-Walk: Learning to walk over graphs using monte carlo tree searchNeurIPS2018link
RARLRule-aware reinforcement learning for knowledge graph reasoningACL-IJCNLP2021link
AttnPathIncorporating graph attention mechanism into knowledge graph reasoning based on deep reinforcement learningEMNLP-IJCNLP2019link
DIVINEDIVINE: A generative adversarial imitation learning framework for knowledge graph reasoningEMNLP-IJCNLP2019link

LLM-based Multi-Step KGR Model

ModelTitleConference/JournalYearPaper
KG&LLM SurveyUnifying large language models and knowledge graphs: A roadmapIEEE TKDE2024link
StructGPTStructGPT: A general framework for large language model to reason over structured dataEMNLP2023link
KSLKnowledge solver: Teaching LLMs to search for domain knowledge from knowledge graphsarXiv2023link
KD-CoTKnowledge-driven CoT: Exploring faithful reasoning in LLMs for knowledge-intensive question answeringarXiv2023link
ToGThink-on-Graph: Deep and responsible reasoning of large language model on knowledge graphICLR2024link
KnowledgeNavigatorKnowledgeNavigator: Leveraging large language models for enhanced reasoning over knowledge graphComplex Intell. Syst.2024link
Nguyen et al.Direct evaluation of chain-of-thought in multi-hop reasoning with knowledge graphsFindings of ACL2024link
KG-AgentKG-Agent: An efficient autonomous agent framework for complex reasoning over knowledge grapharXiv2024link
AgentTuningAgentTuning: Enabling generalized agent abilities for LLMsFindings of ACL2024link
GlamGlam: Fine-tuning large language models for domain knowledge graph alignment via neighborhood partitioning and generative subgraph encodingAAAI Symposium2024link
Taxonomy of static multi-step KGR approaches

Multi-step KGR

⬆️

Dynamic KGR

The illustration of the dynamic KGR task

Dynamic KGR

Incremental KGE Model

ModelTitleConference/JournalYearPaper
DKGEEfficiently embedding dynamic knowledge graphsKnowl.-Based Syst.2022link
PuTransENon-parametric estimation of multiple embeddings for link prediction on dynamic knowledge graphsAAAI2017link
Liu et al.Heuristic-driven, type-specific embedding in parallel spaces for enhancing knowledge graph reasoningICASSP2024link
ABIEAnchors-based incremental embedding for growing knowledge graphsTKDE2023link
CKGETowards continual knowledge graph embedding via incremental distillationAAAI2024link
LKGELifelong embedding learning and transfer for growing knowledge graphsAAAI2023link
AIRAIR: Adaptive incremental embedding updating for dynamic knowledge graphsDASFAA2023link
TIETIE: A framework for embedding-based incremental temporal knowledge graph completionSIGIR2021link
RotatHIncremental update of knowledge graph embedding by rotating on hyperplanesICWS2021link
MMRotatHKnowledge graph incremental embedding for unseen modalitiesKnowl. Inf. Syst.2023link
DKGEEfficiently embedding dynamic knowledge graphsKnowl.-Based Syst.2022link
NaviDynamic knowledge graph embeddings via local embedding reconstructionsESWC (Satellite)2022link
UOKEOnline updates of knowledge graph embeddingComplex Networks X2021link
KGCRTemporal knowledge graph incremental construction model for recommendationAPWeb-WAIM2020link

⬆️

Temporal KGR Model

Time Embedding-based Models

ModelTitleConference/JournalYearPaper
TA-TransELearning sequence encoders for temporal knowledge graph completionEMNLP2018link
HyTEHyTE: Hyperplane-based temporally aware knowledge graph embeddingEMNLP2018link
TTransEDeriving validity time in knowledge graphWWW2018link
TEROTeRo: A time-aware knowledge graph embedding via temporal rotationCOLING2020link
TDistMultEmbedding models for episodic knowledge graphsJWS2019link
TComplExTensor decompositions for temporal knowledge base completionICLR2020link
SimplEDiachronic embedding for temporal knowledge graph completionAAAI2020link
ATiSETemporal KGC based on time series gaussian embeddingISWC2020link
TARGATTARGAT: A time-aware relational graph attention modelIEEE/ACM TASLP2023link
LCGELogic and commonsense-guided TKGCAAAI2023link

Evolution Learning-based Models

ModelTitleConference/JournalYearPaper
Know-EvolveKnow-evolve: deep temporal reasoning for dynamic knowledge graphsICML2017link
RE-NETRecurrent event network: autoregressive structure inference over temporal knowledge graphsEMNLP2020link
EvolveRGCNEvolveGCN: Evolving Graph Convolutional Networks for Dynamic GraphsAAAI2020link
CyGNetLearning from history: modeling temporal knowledge graphs with sequential copy-generation networksAAAI2021link
CluSTeRSearch from history and reason for future: two-stage reasoning on temporal knowledge graphsACL2021link

Temporal Rule Learning

ModelTitleConference/JournalYearPaper
StreamLearnerLearning temporal rules from knowledge graph streamsAAAI Spring Symposium2019link
TlogicTlogic: temporal logical rules for explainable link forecasting on temporal knowledge graphsAAAI2022link
TILPTILP: differentiable learning of temporal logical rules on knowledge graphsICLR2023link
TEILPTEILP: time prediction over knowledge graphs via logical reasoningAAAI2024link
NeuSTIPNeuSTIP: a neuro-symbolic model for link and time prediction in temporal knowledge graphsEMNLP2023link

Multi-step Temporal KGR Model

ModelTitleConference/JournalYearPaper
xERTEExplainable subgraph reasoning for forecasting on temporal knowledge graphsICLR2021link
CluSTeRSearch from history and reason for future: two-stage reasoning on temporal knowledge graphsACL2021link
TPathMulti-hop reasoning over paths in temporal knowledge graphs using reinforcement learningApplied Soft Computing2021link
T-GAPLearning to walk across time for interpretable temporal knowledge graph completionKDD2021link
RTTIReinforcement learning with time intervals for temporal knowledge graph reasoningInformation Systems2024link
TITerTimeTraveler: Reinforcement learning for temporal knowledge graph forecastingEMNLP2021link

LLM-based Temporal KGR Model

ModelTitleConference/JournalYearPaper
PPTPre-trained language model with prompts for temporal knowledge graph completionFindings of ACL2023link
ECOLAECOLA: Enhancing temporal knowledge embeddings with contextualized language representationsFindings of ACL2023link
SToKELearning joint structural and temporal contextualized knowledge embeddings for temporal knowledge graph completionFindings of ACL2023link
NeoXTemporal knowledge graph forecasting without knowledge using in-context learningEMNLP2023link
CSProm-KGDipping PLMs Sauce: Bridging structure and text for effective knowledge graph completion via conditional soft promptingFindings of ACL2023link
zrLLMzrLLM: Zero-shot relational learning on temporal knowledge graphs with large language modelsNAACL2024link
Taxonomy of dynamic KGR approaches

Dynamic KGR

⬆️

Multi-Modal KGR

Multi-Modal Embedding-based Model

ModelTitleConference/JournalYearPaper
Wang et al.Knowledge graph and text jointly embeddingEMNLP2014link
DKRLRepresentation learning of knowledge graphs with entity descriptionsAAAI2016link
TEKEText-enhanced representation learning for knowledge graphIJCAI2016link
KG-BERTModeling relational data with graph convolutional networksESWC2018Link
SimKGCSimKGC: Simple contrastive knowledge graph completion with pre-trained language modelsACL2022link
StARStructure-augmented text representation learning for efficient knowledge graph completionWWW2021link
IKRLImage-embodied knowledge representation learningIJCAI2017link
TransAEMultimodal data enhanced representation learning for knowledge graphsIJCNN2019link
RSMEIs visual context really helpful for knowledge graph? A representation learning perspectiveACM MM2021link
OTKGEOTKGE: multi-modal knowledge graph embeddings via optimal transportNeurIPS2024link
HRGATHyper-node relational graph attention network for multi-modal knowledge graph completionACM TOMM2023link
MKBEEmbedding multimodal relational data for knowledge base completionEMNLP2018link
MMKGRMMKGR: multi-hop multi-modal knowledge graph reasoningICDE2022link
NativENativE: Multi-modal knowledge graph completion in the wildSIGIR2024link
TransFusionTransFusion: Multi-modal fusion for video tag inference via translation-based knowledge embeddingACM MM2021link
MoSEMoSE: modality split and ensemble for multimodal knowledge graph completionEMNLP2022link
IMFIMF: interactive multimodal fusion model for link predictionWWW2023link
MMRNSRelation-enhanced negative sampling for multimodal knowledge graph completionACM MM2022link
MANSModality-aware negative sampling for multi-modal knowledge graph embeddingIJCNN2023link
DHNSDiffusion-based Hierarchical Negative Sampling for Multimodal Knowledge Graph CompletionarXiv2025link

PLM-based Model

ModelTitleConference/JournalYearPaper
VL-BERTVL-BERT: pre-training of generic visual-linguistic representationsICLR2019link
VisualbertVisualbert: A simple and performant baseline for vision and languagearXiv2019link
Unicoder-VLUnicoder-VL: A universal encoder for vision and language by cross-modal pre-trainingAAAI2020link
UNITERUNITER: universal image-text representation learningSpringerLink2020link
LXMERTLXMERT: learning cross-modality encoder representations from transformersEMNLP-IJCNLP2019link
ViLBERTViLBERT: pretraining task-agnostic visiolinguistic representations for vision-and-language tasksNeurIPS2019link
MKGformerHybrid transformer with multi-level fusion for multimodal knowledge graph completionSIGIR2022link
VISTAVISTA: visual-textual knowledge graph representation learningEMNLP Findings2023link
SGMPTStructure guided multi-modal pre-trained transformer for knowledge graph reasoningarXiv2023link
MMKRLMMKRL: a robust embedding approach for multi-modal knowledge graph representation learningApplied Intelligence2022link
KoPAMixture of modality knowledge experts for robust multi-modal knowledge graph completionarXiv2024link
Taxonomy of multi-modal KGR approaches

MM KGR

⬆️

Few-Shot KGR

The illustration of few-shot KGR in the 3-shot setting

FSKGR

Metric Learning-based Model

ModelTitleConference/JournalYearPaper
GMatchingOne-shot relational learning for knowledge graphsEMNLP2018link
FSRLFew-shot knowledge graph completionAAAI2020link
FAANAdaptive attentional network for few-shot knowledge graph completionEMNLP2020link
TransAMExploring entity interactions for few-shot relation learning (student abstract)AAAI2022link
FRL-KGCFew-shot knowledge graph completion model based on relation learningApplied Sciences2023link
HMNetHMNet: hybrid matching network for few-shot link predictionDASFAA2021link
MetapMetap: meta pattern learning for one-shot knowledge graph completionSIGIR2021link

Meta-Learning-based Model

ModelTitleConference/JournalYearPaper
MetaRMeta relational learning for few-shot link prediction in knowledge graphsEMNLP-IJCNLP2019link
GANARelational learning with gated and attentive neighbor aggregator for few-shot knowledge graph completionSIGIR2021link
Meta-iKGSubgraph-aware few-shot inductive link prediction via meta-learningIEEE TKDE2022link
SMetaRSimple and effective meta relational learning for few-shot knowledge graph completionOptimization and Engineering2024link
HiReHierarchical relational learning for few-shot knowledge graph completionarXiv2022link
MTRNTask-related network based on meta-learning for few-shot knowledge graph completionApplied Intelligence2024link

Auxiliary Information-Enhanced Model

ModelTitleConference/JournalYearPaper
TCVAETackling long-tailed relations and uncommon entities in knowledge graph completionEMNLP-IJCNLP2019link
ZSGANGenerative adversarial zero-shot relational learning for knowledge graphsAAAI2020link
HAPZSLHAPZSL: a hybrid attention prototype network for knowledge graph zero-shot relational learningNeurocomputing2022link
OntoZSLOntoZSL: ontology-enhanced zero-shot learningWWW2021link
DOZSLDisentangled ontology embedding for zero-shot learningIJCAI2018link
DMoGDecoupling mixture-of-graphs: unseen relational learning for knowledge graph completion by fusing ontology and textual expertsCOLING2022link
P-INTP-INT: a path-based interaction model for few-shot knowledge graph completionEMNLP Findings2021link
EPIRLEnhancing path information with reinforcement learning for few-shot knowledge graph completionICPADS2023link

⬆️

Multi-Step Few-Shot KGR Model

ModelTitleConference/JournalYearPaper
Meta-KGRAdapting meta knowledge graph information for multi-hop reasoning over few-shot relationsEMNLP-IJCNLP2019link
FIREFew-shot multi-hop relation reasoning over knowledge basesEMNLP2020link
ADK-KGAdapting distilled knowledge for few-shot relation reasoning over knowledge graphsSDM2022link
THMLWhen hardness makes a difference: multi-hop knowledge graph reasoning over few-shot relationsCIKM2021link

Temporal Few-Shot KGR Model

ModelTitleConference/JournalYearPaper
FTMOFew-shot temporal knowledge graph completion based on meta-optimizationComplex Intell. Syst.2023link
TFSCFew-shot link prediction for temporal knowledge graphs based on time-aware translation and attention mechanismNeural Networks2023link
TR-MatchTemporal-relational matching network for few-shot temporal knowledge graph completionDASFAA 20232023link
FTMFFTMF: few-shot temporal knowledge graph completion based on meta-optimization and fault-tolerant mechanismWorld Wide Web2023link
MetaRTFew-shot link prediction with meta-learning for temporal knowledge graphsJ. Comput. Des. Eng.2023link
MetaTKGRLearning to sample and aggregate: few-shot reasoning over temporal knowledge graphsNeurIPS2022link
FITCARLImproving few-shot inductive learning on temporal knowledge graphs using confidence-augmented reinforcement learningMachine Learning and Knowledge Discovery in Databases2023link
Taxonomy of few-shot KGR approaches

FSKGR

⬆️

Inductive KGR

The illustration of inductive KGR

IKGR

Rule-based Model

ModelTitleConference/JournalYearPaper
GraphSAGEInductive representation learning on large graphsNeurIPS2017link
RuleNetMissing-edge aware knowledge graph inductive inference through dual graph learning and traversingExpert Systems with Applications2023link
CBGNNCycle representation learning for inductive relation predictionICML2022link
RED-GNNKnowledge graph reasoning with relational digraphACM Web Conference2022link
VNVN network: embedding newly emerging entities with virtual neighborsCIKM2020link
ARGCNInductive knowledge graph reasoning for multi-batch emerging entitiesCIKM2022link
ELPEExplainable link prediction for emerging entities in knowledge graphsISWC2020link

GNN-based Model

ModelTitleConference/JournalYearPaper
MEANKnowledge transfer for out-of-knowledge-base entities: a graph neural network approachIJCAI2017link
NBFNetNeural Bellman-Ford networks: a general graph neural network framework for link predictionNeurIPS2024link
GraILInductive relation prediction by subgraph reasoningICML2020link
PathConRelational message passing for knowledge graph completionKDD2021link
SNRISubgraph neighboring relations infomax for inductive link prediction on knowledge graphsIJCAI2022link
REPORTInductive relation prediction from relational paths and context with hierarchical transformersICASSP2023link
LogCoInductive relation prediction with logical reasoning using contrastive representationsEMNLP2022link
RPC-IRLearning first-order rules with relational path contrast for inductive relation reasoningarXiv2021link
TACTTopology-aware correlations between relations for inductive link prediction in knowledge graphsAAAI2021link
NRTGEntity representation by neighboring relations topology for inductive relation predictionPRICAI2022link
CoMPILECommunicative message passing for inductive relation reasoningAAAI2021link
LCILPLocality-aware subgraphs for inductive link prediction in knowledge graphsPattern Recognition Letters2023link
ReCoLeRelation-dependent contrastive learning with cluster sampling for inductive relation predictionNeurocomputing2024link
DEKG-ILPDisconnected emerging knowledge graph oriented inductive link predictionICDE2023link
CG-AGGExploring relational semantics for inductive knowledge graph completionAAAI2022link
FCLEntity-AttAttention-based aggregation graph networks for knowledge graph information transferPAKDD2020link
SAGNNOpen-world relationship predictionICTAI2020link
LANLogic attention based neighborhood aggregation for inductive knowledge graph embeddingAAAI2019link
SLANSLAN: similarity-aware aggregation network for embedding out-of-knowledge-graph entitiesNeurocomputing2022link
ARPAttention-based relation prediction of knowledge graph by incorporating graph and context featuresWISE2022link
TransNSOpen knowledge graph representation learning based on neighbors and semantic affinityJournal of Computer Research and Development2019link

⬆️

Multimodal-Enhanced Model

ModelTitleConference/JournalYearPaper
CatEOntological concept structure aware knowledge transfer for inductive knowledge graph embeddingIJCNN2021link
DKRLRepresentation learning of knowledge graphs with entity descriptionsAAAI2016link
OWEAn open-world extension to knowledge graph completion modelsAAAI2019link
WOWEWeighted aggregator for the open-world knowledge graph completionCCIS2020link
Caps-OWKGCaps-OWKG: a capsule network model for open-world knowledge graphInt. J. Mach. Learn. & Cyber.2021link
OWE-MRCExtracting short entity descriptions for open-world extension to knowledge graph completion modelsAdvances in Knowledge Science and Engineering2021link
OWE-RSTRelation specific transformations for open world knowledge graph completionTextGraphs @ ACL2020link
EmReCoEmbeddings based on relation-specific constraints for open world knowledge graph completionApplied Intelligence2023link
ConMaskOpen-world knowledge graph completionAAAI2018link
SDTSDT: an integrated model for open-world knowledge graph reasoningExpert Systems with Applications2020link
Bi-LinkBi-Link: bridging inductive link predictions from text via contrastive learning of transformers and promptsarXiv2022link
RAILDRAILD: towards leveraging relation features for inductive link prediction in knowledge graphsIJCKG2023link
DMoGDecoupling mixture-of-graphs: unseen relational learning for knowledge graph completion by fusing ontology and textual expertsCOLING2022link
BERTRLInductive relation prediction by BERTAAAI2022link
InductivEInductive learning on commonsense knowledge graph completionIJCNN2021link
FITCARLImproving few-shot inductive learning on temporal knowledge graphs using confidence-augmented reinforcement learningMachine Learning and Knowledge Discovery in Databases2023link
TITerTimeTraveler: Reinforcement learning for temporal knowledge graph forecastingEMNLP2021link
MetaTKGRLearning to sample and aggregate: few-shot reasoning over temporal knowledge graphsNeurIPS2022link
FILTFew-shot inductive learning on temporal knowledge graphs using concept-aware informationAKBC2022link
Taxonomy of inductive KGR approaches

IKGR

⬆️

Benchmarks

Datasets for Static KGR Tasks

Dataset#Entities#Relations#Training Triples#Valid Triples#Test TriplesPaper Link
Countries27121,1102424link
Kinship104258,5441,0681,074link
FB1375,04313316,23211,81647,464link
FB1229,73812291,6389,59511,243link
FB15K14,9511,345483,14250,00059,071link
FB15K23714,505237272,11517,53520,466link
FB20K19,9231,452378,07289,04090,143link
FB5M5,385,3221,19219,193,55650,00050,000link
WN1138,58811110,3615,21221,035link
WN1840,94318141,4425,0005,000link
WN18RR40,5591186,8352,9242,924link
YAGO3-10123,143371,079,0404,9784,982link
YAGO37123,18937420,62350,00050,000link
NELL-99575,492200126,1765,0005,000link

Datasets for Dynamic KGR Tasks

Dataset#Entities#RelationsTemporal#Training#Valid#TestPaper Link
GDELT7,691240Timestemp1,033,270238,765305,241link
ICEWS146,738235Timestemp118,76614,85914,756link
ICEWS05-1510,488251Timestemp386,96246,09246,275link
Wikidata12k12,55424Time Interval2,735,685341,961341,961link
YAGO11k10,62310Time Interval161,54019,52320,026link
YAGO15k15,40334Time Interval110,44113,81513,800link

Datasets for Multi-modal KGR Tasks

Dataset#Entities#RelationsModality#Training#Valid#TestPaper Link
FB-IMG-TXT11,7571,231Image+Text285,85034,86329,580link
FB15K237-IMG14,541237Image272,11517,53520,466link
WN9-IMG-TXT6,5559Image+Text11,7411,3191,337link
WN18-IMG40,94318Image141,4425,0005,000link
MKG-Wikipedia15,000169Image34,1964,2744,276link
MKG-YAGO15,00028Image21,3102,6632,665link
TIVA11,85816Video20,0712,0002,000link

Datasets for Few-shot KGR Tasks

Dataset#Entities#Relations#Triples#Training/Valid/Test SplitsPaper Link
NELL-One68,545358181,10951/5/1link
Wiki-One4,868,2448225,859,240133/16/34link
FB15K-One14,541231281,62475/11/33link

Datasets for Inductive KGR Tasks

DatasetVersionTraining/Test Set#Entities#Relations#TriplesPaper Link
FB15K237v1Train2,0001835,226link
Test1,5001462,404
v2Train3,00020312,085
Test2,0001765,092
v3Train4,00021822,394
Test3,0001879,137
v4Train5,00022233,916
Test3,50020414,554
WN18RRv1Train2,74696,678link
Test92291,991
v2Train6,9541018,968
Test2,923104,863
v3Train12,0781132,150
Test5,084117,470
v4Train3,86199,842
Test7,208915,157
NELL-995v1Train10,915145,540link
Test225141,034
v2Train2,5648810,109
Test4,937795,521
v3Train4,64714220,117
Test4,9211229,668
v4Train2,092779,289
Test3,294618,520
Dataset#Entities#Relations#Training Triples#Test TriplesPaper Link
DBPedia50k24,62435132,3886,459link
Wikidata5M4,579,60982220,496,5146,894link

⬆️

Applications

Question Answering

Illustrative examples of the KGR technique applied to QA systems

QA

ModelTitleConference/JournalYearPaper
KBQA SurveyA survey: complex knowledge base question answeringIEEE ICICSE2022link
KEQAKnowledge graph embedding based question answeringACM WSDM2019link
TRL-KEQAQuestion answering over knowledge base embeddings with triples representation learningNeural Information Processing2021link
TransE-QAKnowledge base question answering system based on knowledge graph representation learningACM ICIAI2020link
CAPKGQAComplex question answering over incomplete knowledge graph as n-ary link predictionIEEE IJCNN2022link
EmbedKGQAImproving multi-hop question answering over knowledge graphs using knowledge base embeddingsACL2020link
PKEEQAPath-enhanced multi-relational question answering with knowledge graph embeddingsarXiv2021link
PA-KGQAPath-aware multi-hop question answering over knowledge graph embeddingIEEE ICTAI2022link
HamQAHierarchy-aware multi-hop question answering over knowledge graphsACM Web Conference2023link
BRGNNQuery path generation via bidirectional reasoning for multihop question answering from knowledge basesIEEE TCDS2023link
GRRNImplicit relation inference with deep path extraction for commonsense question answeringNeural Processing Letters2022link
Li et al.Translational relation embeddings for multi-hop knowledge base question answeringWeb Semantics2022link
DSSAGNKnowledge graph multi-hop question answering based on dependent syntactic semantic augmented graph networksElectronics2024link
Jiao et al.A relation embedding assistance networks for multi-hop question answeringACM TALIP2024link
Zhou et al.Marie and BERT – a knowledge graph embedding based question answering system for chemistryACS Omega2023link
CF-KGQACausality-aware enhanced model for multi-hop question answering over knowledge graphsKnowledge-Based Systems2022link
TwiRGCNTwiRGCN: Temporally Weighted Graph Convolution for Question Answering over Temporal Knowledge GraphsEACL2023link
CRONKGQAQuestion answering over temporal knowledge graphsACL-IJCNLP2021link
TempoQRTempoQR: Temporal question reasoning over knowledge graphsAAAI2021link
CTRNAn improving reasoning network for complex question answering over temporal knowledge graphsApplied Intelligence2022link
EXAQTComplex temporal question answering on knowledge graphsACM CIKM2021link
GATQRTemporal knowledge graph question answering models enhanced with GATIEEE BigData2023link
Prog-TQASelf-improvement programming for temporal knowledge graph question answeringLREC-COLING2024link
GenTKGQATwo-stage generative question answering on temporal knowledge graph using large language modelsACL Findings2024link

⬆️

Recommendation

ModelTitleConference/JournalYearPaper
KGCNKnowledge graph convolutional networks for recommender systemsWWW2019link
KGNCF-RRNNeural collaborative recommendation with knowledge graphIEEE ICKG2020link
KGECFKnowledge graph embedding based collaborative filteringIEEE Access2020link
SurveyA review of explainable recommender systems utilizing knowledge graphs and reinforcement learningIEEE Access2024link
PGPRReinforcement knowledge graph reasoning for explainable recommendationSIGIR2019link
CogERCognition-aware knowledge graph reasoning for explainable recommendationWSDM2023link
Hsu et al.Explainable mutual fund recommendation system developed based on knowledge graph embeddingsApplied Intelligence2022link
Lee et al.GCN-based explainable recommendation using a knowledge graph and a language modelIEEE BigData2023link
Markchom et al.Explainable meta-path based recommender systemsACM TORS2023link
Fu et al.Fairness-aware explainable recommendation over knowledge graphsSIGIR2020link
KRRLKnowledge-aware reasoning with self-supervised reinforcement learning for explainable recommendation in MOOCsNeural Computing and Applications2024link
Ryotaro et al.An explainable recommendation framework based on an improved knowledge graph attention network with massive volumes of side informationKnowledge-Based Systems2022link
RippleNetRippleNet: propagating user preferences on the knowledge graph for recommender systemsCIKM2018link
AKUPMAKUPM: attention-enhanced knowledge-aware user preference model for recommendationKDD2019link
RCoLMUnifying task-oriented knowledge graph learning and recommendationIEEE Access2019link
KGATKGAT: knowledge graph attention network for recommendationKDD2019link
IntentGCIntentGC: a scalable graph convolution framework fusing heterogeneous information for recommendationKDD2019link
AKGEHierarchical attentive knowledge graph embedding for personalized recommendationElectronic Commerce Research and Applications2021link
KPRNExplainable reasoning over knowledge graphs for recommendationAAAI2019link

⬆️

Visual Reasoning

Visual Question Answering

ModelTitleConference/JournalYearPaper
FVQAFVQA: fact-based visual question answeringIEEE TPAMI2018link
Wang et al.Explicit knowledge-based reasoning for visual question answeringIJCAI2017link
GraphhopperGraphhopper: multi-hop scene graph reasoning for visual question answeringISWC2021link
Hypergraph TransformerHypergraph transformer: weakly-supervised multi-hop reasoning for knowledge-based visual question answeringACL2022link
CMRLCross-modality multiple relations learning for knowledge-based visual question answeringACM TOMM2024link
KRISPKRISP: Integrating implicit and symbolic knowledge for open-domain knowledge-based VQACVPR2021link
LLM+(KBret+SGret)Find the gap: knowledge base reasoning for visual question answeringarXiv2024link

Cross-Modal Retrieval

ModelTitleConference/JournalYearPaper
KCRKnowledge-aware cross-modal text-image retrieval for remote sensing imagesIEEE TGRS2022link
MMRGMulti-modal relational graph for cross-modal video moment retrievalCVPR2021link
IRGRMultiple instance relation graph reasoning for cross-modal hash retrievalKnowledge-Based Systems2022link

Scene Graph Generation

ModelTitleConference/JournalYearPaper
GB-NetBridging knowledge graphs to generate scene graphsECCV2020link
HiKER-SGGHiKER-SGG: Hierarchical knowledge enhanced robust scene graph generationCVPR2024link
CGRConfigurable graph reasoning for visual relationship detectionTNNLS2022link
COACHERZero-shot scene graph relation prediction through commonsense knowledge integrationECML PKDD2021link

⬆️

Healthcare Domain

ModelTitleConference/JournalYearPaper
Zhu et al.Multimodal reasoning based on knowledge graph embedding for specific diseasesBioinformatics2022link
Chai et al.Diagnosis method of thyroid disease combining knowledge graph and deep learningIEEE Access2020link
SSI-DDISSI-DDI: substructure-substructure interactions for drug-drug interaction predictionBrief. Bioinform.2021link
KGNNKGNN: knowledge graph neural network for drug-drug interaction predictionIJCAI2020link
SMRSMR: medical knowledge graph embedding for safe medicine recommendationBig Data Res.2021link
PharmKGPharmKG: a dedicated knowledge graph benchmark for biomedical data miningBrief. Bioinform.2021link
KG-PredictKG-Predict: a knowledge graph computational framework for drug repurposingJ. Biomed. Inform.2022link

Business Domain

ModelTitleConference/JournalYearPaper
OpenBGConstruction and applications of billion-scale pre-trained multimodal business knowledge graphICDE2023link
Zhang et al.Knowledge graph embedding in e-commerce applications: attentive reasoning, explanations, and transferable rulesInt. Joint Conf. on Knowledge Graphs2021link
Yang et al.Inferring substitutable and complementary products with knowledge-aware path reasoning based on dynamic policy networkKnowledge-Based Syst.2022link
Mitropoulou et al.Anomaly detection in cloud computing using knowledge graph embedding and machine learning mechanismsJ. Grid Comput.2024link
Kosasih et al.Towards knowledge graph reasoning for supply chain risk management using graph neural networksInt. J. Prod. Res.2022link
Yang et al.Research on enterprise risk knowledge graph based on multi-source data fusionNeural Comput. Appl.2022link
Zhang et al.Billion-scale pre-trained e-commerce product knowledge graph modelICDE2021link

Cybersecurity Domain

ModelTitleConference/JournalYearPaper
SikosCybersecurity knowledge graphsKnowl. Inf. Syst.2023link
Ezekia Gilliard et al.Cybersecurity knowledge graph enabled attack chain detection for cyber-physical systemsComputers and Electrical Engineering2023link
Hu et al.Knowledge graph reasoning for cyber attack detectionIET Commun.2024link

Other Domain

ModelTitleConference/JournalYearPaper
Liang et al.Graph path fusion and reinforcement reasoning for recommendation in MOOCsEduc. Inf. Technol.2023link
Zhou et al.Mining tourist preferences and decision support via tourism-oriented knowledge graphInf. Process. Manag.2024link
Gao et al.Hierarchical knowledge graph learning enabled socioeconomic indicator prediction in location-based social networkThe Web Conference (WWW)2023link
Zeng et al.Combining knowledge graph into metro passenger flow prediction: a split-attention relational graph convolutional networkExpert Syst. Appl.2023link
Liu et al.Multi-source knowledge graph reasoning for ocean oil spill detection from satellite SAR imagesInt. J. Appl. Earth Obs. Geoinf.2023link

⬆️

Challenge and Opportunity

Sparse KGR

ModelTitleConference/JournalYearPaper
HoGRNHoGRN: explainable sparse knowledge graph completion via high-order graph reasoning networkIEEE Trans. on Knowledge and Data Engineering2024link
Jia et al.Application of graph neural network and feature information enhancement in relation inference of sparse knowledge graphJournal of Electronic Science and Technology2023link
KRACLKRACL: contrastive learning with graph context modeling for sparse knowledge graph completionThe Web Conference (WWW)2023link
BERT-ConvEEffective use of BERT in graph embeddings for sparse knowledge graph completionACM/SIGAPP Symposium on Applied Computing (SAC)2022link
DacKGRDynamic anticipation and completion for multi-hop reasoning over sparse knowledge graphEMNLP2020link
RuMER-RLRuMER-RL: a hybrid framework for sparse knowledge graph explainable reasoningInformation Sciences2024link
WARWalk-and-relate: a random-walk-based algorithm for representation learning on sparse knowledge graphsarXiv preprint2022link

Uncertain KGR

ModelTitleConference/JournalYearPaper
BEUrREProbabilistic box embeddings for uncertain knowledge graph reasoningNAACL-HLT2021link
SUKESUKE: embedding model for prediction in uncertain knowledge graphIEEE Access2021link
MUKGEEmbedding uncertain knowledge graphsAAAI2019link
UKRMUncertain knowledge graph completion with rule miningWeb Information Systems and Applications2024link
TensorLogTensorlog: a differentiable deductive databasearXiv preprint2016link

KG Error Detection

ModelTitleConference/JournalYearPaper
CKG-EDContrastive knowledge graph error detectionCIKM2022link
CAGEDWhat is normal, what is strange, and what is missing in a knowledge graph: unified characterization via inductive summarizationWWW2020link
HEARKnowledge graph error detection with hierarchical path structureCIKM2023link

Trustworthy KGR

ModelTitleConference/JournalYearPaper
SurveyLogical rule-based knowledge graph reasoning: a comprehensive surveyMathematics2023link
Power-LinkPath-based explanation for knowledge graph completionKDD2024link
IterEIteratively learning embeddings and rules for knowledge graph reasoningWWW2019link
EngineKGPerform like an engine: A closed-loop neural-symbolic learning framework for knowledge graph inferenceCOLING2022link
StreamLearnerLearning temporal rules from knowledge graph streamsAAAI Spring Symposium2019link
TlogicTlogic: temporal logical rules for explainable link forecasting on temporal knowledge graphsAAAI2022link
LCGELogic and commonsense-guided TKGCAAAI2023link
TILPTILP: differentiable learning of temporal logical rules on knowledge graphsICLR2023link
Xu et al.A human-centric evaluation platform for explainable knowledge graph completionEACL (System Demonstrations)2024link
RLF-KGAdvancing Abductive Reasoning in Knowledge Graphs through Complex Logical Hypothesis GenerationACL (Volume 1: Long Papers)2024lihk

LLM-enhanced KGR

ModelTitleConference/JournalYearPaper
KG-GPTKG-GPT: a general framework for reasoning on knowledge graphs using large language modelsFindings of EMNLP2023link
MPIKGCMulti-perspective improvement of knowledge graph completion with large language modelsLREC-COLING2024lihk
LARKComplex logical reasoning over knowledge graphs using large language modelsarXiv2023link
ChatruleChatrule: mining logical rules with large language models for knowledge graph reasoningarXiv2023link
LLM-DALarge language models-guided dynamic adaptation for temporal knowledge graph reasoningarXiv2024link
Xia et al.Chain-of-history reasoning for temporal knowledge graph forecastingFindings of ACL2024link
Luo et al.Chain of history: learning and forecasting with LLMs for temporal knowledge graph completionarXiv2024link
Nguyen et al.Direct evaluation of chain-of-thought in multi-hop reasoning with knowledge graphsFindings of ACL2024link
GenTKGGenTKG: Generative Forecasting on Temporal Knowledge Graph with Large Language ModelsFindings of NAACL2024link

⬆️

awesome
few-shot-learning
inductive-reasoning
knowledge-graph
knowledge-graph-completion
knowledge-graph-embedding
knowledge-graph-reasoning
large-language-model
multi-modal-knowledge-graph
multi-step-reasoning
survey
temporal-reasoning

Contributors

ngl567

34 commits

Similar repos

zjukg/KG-LLM-Papers

[Paper List] Papers integrating knowledge graphs (KGs) and large language models (LLMs)

2,233

XiaoxinHe/Awesome-Graph-LLM

A collection of AWESOME things about Graph-Related LLMs.

2,447

PeterGriffinJin/Awesome-Language-Model-on-Graphs

A curated list of papers and resources based on "Large Language Models on Graphs: A Comprehensive…

996

pengboci/GraphRAG-Survey

290

Graph-RAG/GraphRAG

598

DEEP-PolyU/Awesome-GraphRAG

Awesome-GraphRAG: A curated list of resources (surveys, papers, benchmarks, and opensource…

2,658

woojeongjin/dynamic-KG

Dynamic (Temporal) Knowledge Graph Completion (Reasoning)

610

USTCAGI/Awesome-Papers-Retrieval-Augmented-Generation

89

ngl567/KGR-Survey

A Survey of Task-Oriented Knowledge Graph Reasoning: Status, Applications, and Prospects

81

34 commits

updated Mar 11, 2026

arxiv.org/abs/2506.11012

See the code

README

KGR-Survey

Awesome License: MIT preprint

A Survey of Task-Oriented Knowledge Graph Reasoning: Status, Applications, and Prospects Paper

🚀Main contribution of this survey: This survey provides a more comprehensive perspective on the research of KGR by categorizing approaches based on primary reasoning tasks, downstream application tasks, and potential challenging reasoning tasks. Besides, we explore advanced techniques, such as large language models (LLMs), and their impact on KGR. This work aims to highlight key research trends and outline promising future directions in the field of KGR.

🙌Key characteristics of this repository: Unlike other outstanding review repositories of the knowledge graph reasoning field, we not only provide a comprehensive review but also strive to offer the official publication abstract page for each paper. This includes not only the official publication version of the paper but also additional resources such as author information, videos, datasets, supplementary materials, and BibTeX citations.

If this repository is useful for you, please kindly cite the corresponding survey paper:

@misc{niu2025kgrsurvey,
  author       = {Guanglin Niu and Bo Li and Yangguang Lin},
  title        = {A Survey of Task‐Oriented Knowledge Graph Reasoning: Status, Applications, and Prospects},
  year         = {2025},
  eprint       = {arXiv:2506.11012},
  archivePrefix= {arXiv},
  primaryClass = {cs.AI},
  url          = {https://arxiv.org/abs/2506.11012}
}

The comprehensive overview framework of our survey is presented as following. The same number (①-⑨) indicates that different approaches share similar ideas, and the keywords corresponding to each number are provided at the bottom of the figure.

The illustration of the six primary KGR tasks

Illustration of primary KGR tasks

🔥 News

Content

Survey Papers

TitleConference/JournalYearCharacteristicPaper
A survey of task-oriented knowledge graph reasoning: status, applications, and prospectsarXiv2025Task-oriented KGRlink
Knowledge graph embedding: a survey from the perspective of representation spacesACM Computer Survey2024Embedding Spaceslink
A survey of knowledge graph reasoning on graph types: Static, dynamic, and multi-modalIEEE TPAMI2024Graph Typeslink
Negative sampling in knowledge graph representation learning: a reviewarXiv2024Negative Samplinglink
Overview of knowledge reasoning for knowledge graphNeurocomputering2024Causal Reasoninglink
A survey on temporal knowledge graph: representation learning and applicationsarXiv2024Temporal Reasoninglink
A survey on temporal knowledge graph completion: taxonomy, progress, and prospectsarXiv2023Temporal Reasoninglink
Generalizing to unseen elements: a survey on knowledge extrapolation for knowledge graphsIJCAI2023Unseen Elementslink
A survey on few-shot knowledge graph completion with structural and commonsense knowledgearXiv2023Commonsenselink
Beyond transduction: a survey on inductive, few shot, and zero shot link prediction in knowledge graphsarXiv2023Few-shot & Inductivelink
A comprehensive overview of knowledge graph completionKnowledge-Based System2022Multi-modal & Hyper-relationlink
Knowledgegraph reasoning with logics and embeddings: survey and perspectivearXiv2022Logics and Embeddingslink

⬆️

Static Single-Step KGR

KGE-based KGR Model

The illustration of five representative KGE models

Illustration of primary KGR tasks

Translation or Tensor Decomposition-Based KGE Models

ModelTitleConference/JournalYearPaper
TransETranslating embeddings for modeling multi-relational dataNIPS2013link
TransHKnowledge graph embedding by translating on hyperplanesAAAI2014link
TransRLearning entity and relation embeddings for knowledge graph completionAAAI2015link
TransDKnowledge graph embedding via dynamic mapping matrixACL2015link
TranSparseKnowledge graph completion with adaptive sparse transfer matrixAAAI2016link
PairEPairRE: Knowledge graph embeddings via paired relation vectorsACL2021link
TransATransA: An adaptive approach for knowledge graph embeddingarXiv2015link
KG2ELearning to represent knowledge graphs with Gaussian embeddingCIKM2015link
ManifoldEFrom one point to a manifold: Knowledge graph embedding for precise link predictionIJCAI2016link
TorusETorusE: Knowledge graph embedding on a Lie groupAAAI2018link
PoincaréPoincare embeddings for learning hierarchical representationsNIPS2017link
MuRPMulti-relational Poincare graph embeddingsNIPS2019link
HAKELearning hierarchy-aware knowledge graph embeddings for link predictionAAAI2020link
H2EKnowledge graph representation via hierarchical hyperbolic neural graph embeddingIEEE Big Data2021link
HBEHyperbolic hierarchy-aware knowledge graph embedding for link predictionEMNLP2021link
RotatERotatE: Knowledge graph embedding by relational rotation in complex spaceICLR2019link
QuatEQuaternion knowledge graph embeddingNIPS2019link
DualEDual quaternion knowledge graph embeddingsAAAI2021link
RESCALA three-way model for collective learning on multi-relational dataICML2011link
PITF-BPRPredicting RDF triples in incomplete knowledge bases with tensor factorizationSAC2012link
DistMultEmbedding entities and relations for learning and inference in knowledge basesICLR2015link
ComplExComplex embeddings for simple link predictionICML2016link
HolEHolographic embeddings of knowledge graphsAAAI2016link

(Graph) Neural Network-based Models

ModelTitleConference/JournalYearPaper
NTNReasoning with neural tensor networks for knowledge base completionNIPS2013link
SMEA semantic matching energy function for learning with multi-relational dataMachine Learning2014link
NAMProbabilistic reasoning via deep learning: Neural association modelsarXiv2016link
ConvEConvolutional 2D knowledge graph embeddingsAAAI2018link
ConvKBA novel embedding model for knowledge base completion based on convolutional neural networkNAACL2018link
GNN SurveyA comprehensive survey on graph neural networksIEEE TNNLS2021link
R-GCNModeling relational data with graph convolutional networksESWC2018Link
SACNEnd-to-end structure-aware convolutional networks for knowledge base completionAAAI2019link
KBGATLearning attention-based embeddings for relation prediction in knowledge graphsACL2019link
KE-GCNKnowledge embedding based graph convolutional networkThe Web Conference2021link

Transformer-based Models

ModelTitleConference/JournalYearPaper
KG-BERTModeling relational data with graph convolutional networksESWC2018Link
R-MeNA relational memory-based embedding model for triple classification and search personalizationACL2021link
CoKECoKE: Contextualized knowledge graph embeddingarXiv2019link
HittERHittER: Hierarchical transformers for knowledge graph embeddingsEMNLP2021link
GenKGCFrom discrimination to generation: Knowledge graph completion with generative transformerWWW2022link
iHTPre-training transformers for knowledge graph completionarXiv2023link
SimKGCSimKGC: Simple contrastive knowledge graph completion with pre-trained language modelsACL2022link
StARStructure-augmented text representation learning for efficient knowledge graph completionWWW2021link
KoPAMaking large language models perform better in knowledge graph completionarXiv2023link
KICGPTKICGPT: Large language model with knowledge in context for knowledge graph completionEMNLP2023link
RelphormerRelphormer: Relational graph transformer for knowledge graph representationsNeurocomputing2024link
LGKGRLGKGR: A knowledge graph reasoning model using LLMs augmented GNNsNeurocomputing2025Link

⬆️

Ontology-Enhanced KGE Models

ModelTitleConference/JournalYearPaper
JOIEUniversal representation learning of knowledge bases by jointly embedding instances and ontological conceptsKDD2019Link
Nickel et al.Factorizing YAGO: Scalable machine learning for linked dataWWW2012link
CISSEmbedding two-view knowledge graphs with class inheritance and structural similarityKDD2024link
Wang et al.An ontology-enhanced knowledge graph embedding methodICCPR2024link
Concept2BoxConcept2Box: Joint geometric embeddings for learning two-view knowledge graphsACL2023link
CAKECAKE: A scalable commonsense-aware framework for multi-view knowledge graph completionACL2022link
SSESemantically smooth knowledge graph embeddingACL2015link
TKRLRepresentation learning of knowledge graphs with hierarchical typesIJCAI2016link
TransETTransET: Knowledge graph embedding with entity typesElectronics2021link
AutoETERAutoETER: Automated entity type representation for knowledge graph embeddingEMNLP2020link

Path-Enhanced KGE Models

ModelTitleConference/JournalYearPaper
Path-RNNCompositional vector space models for knowledge base completionACL2015link
PTransEModeling relation paths for representation learning of knowledge basesEMNLP2015Link
PRNA path-based relation networks model for knowledge graph completionExpert Systems with Applications2021link
OPTransERepresentation learning with ordered relation paths for knowledge graph completionEMNLP-IJCNLP2019link
TransE&RWModeling relation paths for knowledge base completion via joint adversarial trainingKnowledge Based Systems2020link
HARPAHARPA: hierarchical attention with relation paths for knowledge graph embedding adversarial learningData Mining and Knowledge Discovery2023link
RPJERule-guided compositional representation learning on knowledge graphsAAAI2020link
PARLAttention-aware path-based relation extraction for medical knowledge graphSmart Computing and Communication2017link
Das et al.Chains of reasoning over entities, relations, and text using recurrent neural networksEACL2017link
Jiang et al.Attentive path combination for knowledge graph completionMachine Learning Research2017link
CPConvKEA confidence-aware and path-enhanced convolutional neural network embedding framework on noisy knowledge graphNeurocomputing2023link
PaSKoGEPath-specific knowledge graph embeddingKnowledge-based Systems2018link
Jagvaral et al.Path-based reasoning approach for knowledge graph completion using CNN-BiLSTM with attention mechanismExpert Systems with Applications2020link
PathConRelational message passing for knowledge graph completionKDD2021link
PTrustEPTrustE: A high-accuracy knowledge graph noise detection method based on path trustworthiness and triple embeddingKnowledge-based Systems2022link
TAPRModeling relation paths for knowledge graph completionIEEE TKDE2021link
Niu et al.Joint semantics and data-driven path representation for knowledge graph reasoningNeurocomputing2022link

⬆️

Negative Sampling for KGE

The illustration of six types of negative sampling strategies

Illustration of negative sampling strategies

ModelTitleConference/JournalYearPaper
Local Closed-World AssumptionKnowledge Vault: A web scale approach to probabilistic knowledge fusionKDD2014link
NS SurveyNegative sampling in knowledge graph representation learning: A reviewarXiv2023link
Uniform SamplingKnowledge graph embedding by translating on hyperplanesAAAI2014link
KBGANKBGAN: Adversarial learning for knowledge graph embeddingsNAACL2018Link
Self-AdvRotatE: Knowledge graph embedding by relational rotation in complex spaceICLR2019link
Batch NSPytorch-BigGraph: A large scale graph embedding systemMachine Learning and Systems2019link
Bernoulli NSAn interpretable knowledge transfer model for knowledge base completionACL2017link
Zhang et al.A novel negative sample generating method for knowledge graph embeddingEWSN2019link
SparseNSGA novel negative sampling based on frequency of relational association entities for knowledge graph embeddingJournal of Web Engineering2021link
IGANIncorporating GAN for negative sampling in knowledge representation learningAAAI2018link
GraphGANGraphGAN: Graph representation learning with generative adversarial netsAAAI2018link
KSGANA knowledge selective adversarial network for link prediction in knowledge graphNLPCC2019link
RUGAImproving knowledge graph completion using soft rules and adversarial learningChinese Journal of Electronics2021link
LASAdversarial knowledge representation learning without external modelIEEE Access2019link
ASARelation-aware graph attention model with adaptive self-adversarial trainingAAAI2021link
ANKnowledge graph embedding based on adaptive negative samplingICPSEE2019link
EANSEntity aware negative sampling with auxiliary loss of false negative prediction for knowledge graph embeddingarXiv2022link
Truncated NSFusing attribute character embeddings with truncated negative sampling for entity alignmentElectronics2023link
DNSDistributional negative sampling for knowledge base completionarXiv2019link
ESNSEntity similarity-based negative sampling for knowledge graph embeddingPRICAI2022Link
RCWCKGBoost: A classification-based knowledge base completion method with negative samplingPattern Recognition Letters2022link
Conditional SamplingConditional constraints for knowledge graph embeddingsDL4KG2020link
LEMONLEMON: LanguagE MOdel for negative sampling of knowledge graph embeddingsarXiv preprint2022Link
NSCachingNSCaching: Simple and efficient negative sampling for knowledge graph embeddingICDE2019Link
MDNcachingMDNcaching: A strategy to generate quality negatives for knowledge graph embeddingIEA/AIE2022Link
Op-TransOp-Trans: An optimization framework for negative sampling and triplet-mapping properties in knowledge graph embeddingApplied Sciences2023Link
NS-KGEEfficient non-sampling knowledge graph embeddingThe Web Conference2021Link

⬆️

Open-Source Library for KGE

LibraryImplementationKey FeaturesGitHub Repository
OpenKEPytorch, TensorFlow, C++Efficiently implements fundamental operations such as data loading, negative sampling, and performance evaluation using C++ for high performance.https://github.com/thunlp/OpenKE
AmpliGraphTensorFlowProvides a Keras-style API with improved efficiency over OpenKE.https://github.com/Accenture/AmpliGraph
torchKGEPytorchAchieves twice the efficiency of OpenKE and five times that of AmpliGraph.https://github.com/torchkge-team/torchkge
LibKGEPytorchEnables direct configuration of hyperparameters and model settings via configuration files.https://github.com/uma-pi1/kge
KB2EC++One of the earliest KGE libraries and the predecessor of OpenKE.https://github.com/thunlp/KB2E
scikit-kgePythonImplements multiple classical KGE models and supports a novel negative sampling strategy.https://github.com/mnick/scikit-kge
NeuralKGPytorchIntegrates KGE techniques with graph neural networks (GNNs) and rule-based reasoning models.https://github.com/zjukg/NeuralKG
PyKEENPytorchOffers 37 datasets, 40 KGE models, 15 loss functions, 6 regularization mechanisms, and 3 negative sampling strategies.https://github.com/pykeen/pykeen
Pykg2vecPytorch, TensorFlowSupports automated hyperparameter tuning, exports KG embeddings in TSV or RDF formats, and provides visualization for performance evaluation.https://github.com/Sujit-O/pykg2vec
μKGPytorch, TensorFlowSupports multi-process execution and GPU-accelerated computation, making it well-suited for large-scale KGs.https://github.com/nju-websoft/muKG
DGL-KEPytorch, MXNetOptimized for execution on CPU and GPU clusters, offering high scalability for large-scale KGs.https://github.com/awslabs/dgl-ke
GraphVitePytorchProvides efficient large-scale embedding learning, supports visualization of graph data, and enables multi-processing and GPU parallelization.https://github.com/DeepGraphLearning/graphvite
PBGPytorchDesigned for distributed training, capable of handling KGs with billions of entities and trillions of edges.https://github.com/facebookresearch/PyTorch-BigGraph

⬆️

Logic Rule-based KGR Model

Rule Learning for KG

ModelTitleConference/JournalYearPaper
FOILLearning logical definitions from relationsMachine Learning1990link
MDIEInverse entailment and progolNew Generation Computing1995link
InspireBest-effort inductive logic programming via fine-grained cost-based hypothesis generationMachine Learning2018link
Neural-Num-LPDifferentiable learning of numerical rules in knowledge graphsICLR2020link
AMIE+Fast rule mining in ontological knowledge bases with AMIE+VLDB Journal2015link
ScaLeKBScaLeKB: Scalable learning and inference over large knowledge basesVLDB Journal2016link
RDF2rulesRDF2Rules: Learning rules from RDF knowledge bases by mining frequent predicate cyclesarXiv2015link
SWARMSWARM: An approach for mining semantic association rules from semantic web dataPRICAI2016link
RudikRudik: Rule discovery in knowledge basesPVLDB2018link
RuLESRule learning from knowledge graphs guided by embedding modelsESWC2018link
EvodaRule learning over knowledge graphs with genetic logic programmingICDE2022link
NeuralLPDifferentiable learning of logical rules for knowledge base reasoningNeurIPS2017link
DRUMDRUM: End-to-end differentiable rule mining on knowledge graphsNeurIPS2019link
RLvLRAn embedding-based approach to rule learning in knowledge graphsIEEE TKDE2019link
RNNLogicRNNLogic: learning logic rules for reasoning on knowledge graphsICLR2021link
RARLRelatedness and TBox-driven rule learning in large knowledge basesAAAI2020link
RuleformerRuleformer: context-aware rule mining over knowledge graphCOLING2022link
Ott et al.Rule-based knowledge graph completion with canonical modelsCIKM2023link

Neural-Symbolic KGR

ModelTitleConference/JournalYearPaper
KALEJointly embedding knowledge graphs and logical rulesEMNLP2016link
RUGEKnowledge graph embedding with iterative guidance from soft rulesAAAI2018link
RulERulE: Knowledge graph reasoning with rule embeddingFindings of ACL2024link
RPJERule-guided compositional representation learning on knowledge graphsAAAI2020link
IterEIteratively learning embeddings and rules for knowledge graph reasoningWWW2019link
UniKERUniKER: A unified framework for combining embedding and definite Horn rule reasoning for knowledge graph inferenceEMNLP2021link
EngineKGPerform like an engine: A closed-loop neural-symbolic learning framework for knowledge graph inferenceCOLING2022link
Taxonomy of static single-step KGR approaches

Single-step KGR

⬆️

Static Multi-Step KGR

Random Walk-based Model

ModelTitleConference/JournalYearPaper
PRARelational retrieval using a combination of path-constrained random walksMachine Learning2010link
Lao et al. 1Random walk inference and learning in a large scale knowledge baseEMNLP2011link
Lao et al. 2Reading the web with learned syntactic-semantic inference rulesEMNLP2012link
Gardner et al.Improving learning and inference in a large knowledge-base using latent syntactic cuesEMNLP2013link
CPRAKnowledge base completion via coupled path rankingACL2016link
C-PRContext-aware path ranking for knowledge base completionIJCAI2017link
A*NetA*Net: a scalable path-based reasoning approach for knowledge graphsNeurIPS2024link
SFEEfficient and expressive knowledge base completion using subgraph feature extractionEMNLP2015link
PathConRelational message passing for knowledge graph completionKDD2021link

Reinforcement Learning-based Model

ModelTitleConference/JournalYearPaper
DeepPathDeepPath: a reinforcement learning method for knowledge graph reasoningEMNLP2017link
MINERVAGo for a walk and arrive at the answer: Reasoning over paths in knowledge bases using reinforcement learningICLR2018link
DIVAVariational knowledge graph reasoningNAACL2018link
MultiHopKGMulti-hop knowledge graph reasoning with reward shapingEMNLP2018link
M-WalkM-Walk: Learning to walk over graphs using monte carlo tree searchNeurIPS2018link
RARLRule-aware reinforcement learning for knowledge graph reasoningACL-IJCNLP2021link
AttnPathIncorporating graph attention mechanism into knowledge graph reasoning based on deep reinforcement learningEMNLP-IJCNLP2019link
DIVINEDIVINE: A generative adversarial imitation learning framework for knowledge graph reasoningEMNLP-IJCNLP2019link

LLM-based Multi-Step KGR Model

ModelTitleConference/JournalYearPaper
KG&LLM SurveyUnifying large language models and knowledge graphs: A roadmapIEEE TKDE2024link
StructGPTStructGPT: A general framework for large language model to reason over structured dataEMNLP2023link
KSLKnowledge solver: Teaching LLMs to search for domain knowledge from knowledge graphsarXiv2023link
KD-CoTKnowledge-driven CoT: Exploring faithful reasoning in LLMs for knowledge-intensive question answeringarXiv2023link
ToGThink-on-Graph: Deep and responsible reasoning of large language model on knowledge graphICLR2024link
KnowledgeNavigatorKnowledgeNavigator: Leveraging large language models for enhanced reasoning over knowledge graphComplex Intell. Syst.2024link
Nguyen et al.Direct evaluation of chain-of-thought in multi-hop reasoning with knowledge graphsFindings of ACL2024link
KG-AgentKG-Agent: An efficient autonomous agent framework for complex reasoning over knowledge grapharXiv2024link
AgentTuningAgentTuning: Enabling generalized agent abilities for LLMsFindings of ACL2024link
GlamGlam: Fine-tuning large language models for domain knowledge graph alignment via neighborhood partitioning and generative subgraph encodingAAAI Symposium2024link
Taxonomy of static multi-step KGR approaches

Multi-step KGR

⬆️

Dynamic KGR

The illustration of the dynamic KGR task

Dynamic KGR

Incremental KGE Model

ModelTitleConference/JournalYearPaper
DKGEEfficiently embedding dynamic knowledge graphsKnowl.-Based Syst.2022link
PuTransENon-parametric estimation of multiple embeddings for link prediction on dynamic knowledge graphsAAAI2017link
Liu et al.Heuristic-driven, type-specific embedding in parallel spaces for enhancing knowledge graph reasoningICASSP2024link
ABIEAnchors-based incremental embedding for growing knowledge graphsTKDE2023link
CKGETowards continual knowledge graph embedding via incremental distillationAAAI2024link
LKGELifelong embedding learning and transfer for growing knowledge graphsAAAI2023link
AIRAIR: Adaptive incremental embedding updating for dynamic knowledge graphsDASFAA2023link
TIETIE: A framework for embedding-based incremental temporal knowledge graph completionSIGIR2021link
RotatHIncremental update of knowledge graph embedding by rotating on hyperplanesICWS2021link
MMRotatHKnowledge graph incremental embedding for unseen modalitiesKnowl. Inf. Syst.2023link
DKGEEfficiently embedding dynamic knowledge graphsKnowl.-Based Syst.2022link
NaviDynamic knowledge graph embeddings via local embedding reconstructionsESWC (Satellite)2022link
UOKEOnline updates of knowledge graph embeddingComplex Networks X2021link
KGCRTemporal knowledge graph incremental construction model for recommendationAPWeb-WAIM2020link

⬆️

Temporal KGR Model

Time Embedding-based Models

ModelTitleConference/JournalYearPaper
TA-TransELearning sequence encoders for temporal knowledge graph completionEMNLP2018link
HyTEHyTE: Hyperplane-based temporally aware knowledge graph embeddingEMNLP2018link
TTransEDeriving validity time in knowledge graphWWW2018link
TEROTeRo: A time-aware knowledge graph embedding via temporal rotationCOLING2020link
TDistMultEmbedding models for episodic knowledge graphsJWS2019link
TComplExTensor decompositions for temporal knowledge base completionICLR2020link
SimplEDiachronic embedding for temporal knowledge graph completionAAAI2020link
ATiSETemporal KGC based on time series gaussian embeddingISWC2020link
TARGATTARGAT: A time-aware relational graph attention modelIEEE/ACM TASLP2023link
LCGELogic and commonsense-guided TKGCAAAI2023link

Evolution Learning-based Models

ModelTitleConference/JournalYearPaper
Know-EvolveKnow-evolve: deep temporal reasoning for dynamic knowledge graphsICML2017link
RE-NETRecurrent event network: autoregressive structure inference over temporal knowledge graphsEMNLP2020link
EvolveRGCNEvolveGCN: Evolving Graph Convolutional Networks for Dynamic GraphsAAAI2020link
CyGNetLearning from history: modeling temporal knowledge graphs with sequential copy-generation networksAAAI2021link
CluSTeRSearch from history and reason for future: two-stage reasoning on temporal knowledge graphsACL2021link

Temporal Rule Learning

ModelTitleConference/JournalYearPaper
StreamLearnerLearning temporal rules from knowledge graph streamsAAAI Spring Symposium2019link
TlogicTlogic: temporal logical rules for explainable link forecasting on temporal knowledge graphsAAAI2022link
TILPTILP: differentiable learning of temporal logical rules on knowledge graphsICLR2023link
TEILPTEILP: time prediction over knowledge graphs via logical reasoningAAAI2024link
NeuSTIPNeuSTIP: a neuro-symbolic model for link and time prediction in temporal knowledge graphsEMNLP2023link

Multi-step Temporal KGR Model

ModelTitleConference/JournalYearPaper
xERTEExplainable subgraph reasoning for forecasting on temporal knowledge graphsICLR2021link
CluSTeRSearch from history and reason for future: two-stage reasoning on temporal knowledge graphsACL2021link
TPathMulti-hop reasoning over paths in temporal knowledge graphs using reinforcement learningApplied Soft Computing2021link
T-GAPLearning to walk across time for interpretable temporal knowledge graph completionKDD2021link
RTTIReinforcement learning with time intervals for temporal knowledge graph reasoningInformation Systems2024link
TITerTimeTraveler: Reinforcement learning for temporal knowledge graph forecastingEMNLP2021link

LLM-based Temporal KGR Model

ModelTitleConference/JournalYearPaper
PPTPre-trained language model with prompts for temporal knowledge graph completionFindings of ACL2023link
ECOLAECOLA: Enhancing temporal knowledge embeddings with contextualized language representationsFindings of ACL2023link
SToKELearning joint structural and temporal contextualized knowledge embeddings for temporal knowledge graph completionFindings of ACL2023link
NeoXTemporal knowledge graph forecasting without knowledge using in-context learningEMNLP2023link
CSProm-KGDipping PLMs Sauce: Bridging structure and text for effective knowledge graph completion via conditional soft promptingFindings of ACL2023link
zrLLMzrLLM: Zero-shot relational learning on temporal knowledge graphs with large language modelsNAACL2024link
Taxonomy of dynamic KGR approaches

Dynamic KGR

⬆️

Multi-Modal KGR

Multi-Modal Embedding-based Model

ModelTitleConference/JournalYearPaper
Wang et al.Knowledge graph and text jointly embeddingEMNLP2014link
DKRLRepresentation learning of knowledge graphs with entity descriptionsAAAI2016link
TEKEText-enhanced representation learning for knowledge graphIJCAI2016link
KG-BERTModeling relational data with graph convolutional networksESWC2018Link
SimKGCSimKGC: Simple contrastive knowledge graph completion with pre-trained language modelsACL2022link
StARStructure-augmented text representation learning for efficient knowledge graph completionWWW2021link
IKRLImage-embodied knowledge representation learningIJCAI2017link
TransAEMultimodal data enhanced representation learning for knowledge graphsIJCNN2019link
RSMEIs visual context really helpful for knowledge graph? A representation learning perspectiveACM MM2021link
OTKGEOTKGE: multi-modal knowledge graph embeddings via optimal transportNeurIPS2024link
HRGATHyper-node relational graph attention network for multi-modal knowledge graph completionACM TOMM2023link
MKBEEmbedding multimodal relational data for knowledge base completionEMNLP2018link
MMKGRMMKGR: multi-hop multi-modal knowledge graph reasoningICDE2022link
NativENativE: Multi-modal knowledge graph completion in the wildSIGIR2024link
TransFusionTransFusion: Multi-modal fusion for video tag inference via translation-based knowledge embeddingACM MM2021link
MoSEMoSE: modality split and ensemble for multimodal knowledge graph completionEMNLP2022link
IMFIMF: interactive multimodal fusion model for link predictionWWW2023link
MMRNSRelation-enhanced negative sampling for multimodal knowledge graph completionACM MM2022link
MANSModality-aware negative sampling for multi-modal knowledge graph embeddingIJCNN2023link
DHNSDiffusion-based Hierarchical Negative Sampling for Multimodal Knowledge Graph CompletionarXiv2025link

PLM-based Model

ModelTitleConference/JournalYearPaper
VL-BERTVL-BERT: pre-training of generic visual-linguistic representationsICLR2019link
VisualbertVisualbert: A simple and performant baseline for vision and languagearXiv2019link
Unicoder-VLUnicoder-VL: A universal encoder for vision and language by cross-modal pre-trainingAAAI2020link
UNITERUNITER: universal image-text representation learningSpringerLink2020link
LXMERTLXMERT: learning cross-modality encoder representations from transformersEMNLP-IJCNLP2019link
ViLBERTViLBERT: pretraining task-agnostic visiolinguistic representations for vision-and-language tasksNeurIPS2019link
MKGformerHybrid transformer with multi-level fusion for multimodal knowledge graph completionSIGIR2022link
VISTAVISTA: visual-textual knowledge graph representation learningEMNLP Findings2023link
SGMPTStructure guided multi-modal pre-trained transformer for knowledge graph reasoningarXiv2023link
MMKRLMMKRL: a robust embedding approach for multi-modal knowledge graph representation learningApplied Intelligence2022link
KoPAMixture of modality knowledge experts for robust multi-modal knowledge graph completionarXiv2024link
Taxonomy of multi-modal KGR approaches

MM KGR

⬆️

Few-Shot KGR

The illustration of few-shot KGR in the 3-shot setting

FSKGR

Metric Learning-based Model

ModelTitleConference/JournalYearPaper
GMatchingOne-shot relational learning for knowledge graphsEMNLP2018link
FSRLFew-shot knowledge graph completionAAAI2020link
FAANAdaptive attentional network for few-shot knowledge graph completionEMNLP2020link
TransAMExploring entity interactions for few-shot relation learning (student abstract)AAAI2022link
FRL-KGCFew-shot knowledge graph completion model based on relation learningApplied Sciences2023link
HMNetHMNet: hybrid matching network for few-shot link predictionDASFAA2021link
MetapMetap: meta pattern learning for one-shot knowledge graph completionSIGIR2021link

Meta-Learning-based Model

ModelTitleConference/JournalYearPaper
MetaRMeta relational learning for few-shot link prediction in knowledge graphsEMNLP-IJCNLP2019link
GANARelational learning with gated and attentive neighbor aggregator for few-shot knowledge graph completionSIGIR2021link
Meta-iKGSubgraph-aware few-shot inductive link prediction via meta-learningIEEE TKDE2022link
SMetaRSimple and effective meta relational learning for few-shot knowledge graph completionOptimization and Engineering2024link
HiReHierarchical relational learning for few-shot knowledge graph completionarXiv2022link
MTRNTask-related network based on meta-learning for few-shot knowledge graph completionApplied Intelligence2024link

Auxiliary Information-Enhanced Model

ModelTitleConference/JournalYearPaper
TCVAETackling long-tailed relations and uncommon entities in knowledge graph completionEMNLP-IJCNLP2019link
ZSGANGenerative adversarial zero-shot relational learning for knowledge graphsAAAI2020link
HAPZSLHAPZSL: a hybrid attention prototype network for knowledge graph zero-shot relational learningNeurocomputing2022link
OntoZSLOntoZSL: ontology-enhanced zero-shot learningWWW2021link
DOZSLDisentangled ontology embedding for zero-shot learningIJCAI2018link
DMoGDecoupling mixture-of-graphs: unseen relational learning for knowledge graph completion by fusing ontology and textual expertsCOLING2022link
P-INTP-INT: a path-based interaction model for few-shot knowledge graph completionEMNLP Findings2021link
EPIRLEnhancing path information with reinforcement learning for few-shot knowledge graph completionICPADS2023link

⬆️

Multi-Step Few-Shot KGR Model

ModelTitleConference/JournalYearPaper
Meta-KGRAdapting meta knowledge graph information for multi-hop reasoning over few-shot relationsEMNLP-IJCNLP2019link
FIREFew-shot multi-hop relation reasoning over knowledge basesEMNLP2020link
ADK-KGAdapting distilled knowledge for few-shot relation reasoning over knowledge graphsSDM2022link
THMLWhen hardness makes a difference: multi-hop knowledge graph reasoning over few-shot relationsCIKM2021link

Temporal Few-Shot KGR Model

ModelTitleConference/JournalYearPaper
FTMOFew-shot temporal knowledge graph completion based on meta-optimizationComplex Intell. Syst.2023link
TFSCFew-shot link prediction for temporal knowledge graphs based on time-aware translation and attention mechanismNeural Networks2023link
TR-MatchTemporal-relational matching network for few-shot temporal knowledge graph completionDASFAA 20232023link
FTMFFTMF: few-shot temporal knowledge graph completion based on meta-optimization and fault-tolerant mechanismWorld Wide Web2023link
MetaRTFew-shot link prediction with meta-learning for temporal knowledge graphsJ. Comput. Des. Eng.2023link
MetaTKGRLearning to sample and aggregate: few-shot reasoning over temporal knowledge graphsNeurIPS2022link
FITCARLImproving few-shot inductive learning on temporal knowledge graphs using confidence-augmented reinforcement learningMachine Learning and Knowledge Discovery in Databases2023link
Taxonomy of few-shot KGR approaches

FSKGR

⬆️

Inductive KGR

The illustration of inductive KGR

IKGR

Rule-based Model

ModelTitleConference/JournalYearPaper
GraphSAGEInductive representation learning on large graphsNeurIPS2017link
RuleNetMissing-edge aware knowledge graph inductive inference through dual graph learning and traversingExpert Systems with Applications2023link
CBGNNCycle representation learning for inductive relation predictionICML2022link
RED-GNNKnowledge graph reasoning with relational digraphACM Web Conference2022link
VNVN network: embedding newly emerging entities with virtual neighborsCIKM2020link
ARGCNInductive knowledge graph reasoning for multi-batch emerging entitiesCIKM2022link
ELPEExplainable link prediction for emerging entities in knowledge graphsISWC2020link

GNN-based Model

ModelTitleConference/JournalYearPaper
MEANKnowledge transfer for out-of-knowledge-base entities: a graph neural network approachIJCAI2017link
NBFNetNeural Bellman-Ford networks: a general graph neural network framework for link predictionNeurIPS2024link
GraILInductive relation prediction by subgraph reasoningICML2020link
PathConRelational message passing for knowledge graph completionKDD2021link
SNRISubgraph neighboring relations infomax for inductive link prediction on knowledge graphsIJCAI2022link
REPORTInductive relation prediction from relational paths and context with hierarchical transformersICASSP2023link
LogCoInductive relation prediction with logical reasoning using contrastive representationsEMNLP2022link
RPC-IRLearning first-order rules with relational path contrast for inductive relation reasoningarXiv2021link
TACTTopology-aware correlations between relations for inductive link prediction in knowledge graphsAAAI2021link
NRTGEntity representation by neighboring relations topology for inductive relation predictionPRICAI2022link
CoMPILECommunicative message passing for inductive relation reasoningAAAI2021link
LCILPLocality-aware subgraphs for inductive link prediction in knowledge graphsPattern Recognition Letters2023link
ReCoLeRelation-dependent contrastive learning with cluster sampling for inductive relation predictionNeurocomputing2024link
DEKG-ILPDisconnected emerging knowledge graph oriented inductive link predictionICDE2023link
CG-AGGExploring relational semantics for inductive knowledge graph completionAAAI2022link
FCLEntity-AttAttention-based aggregation graph networks for knowledge graph information transferPAKDD2020link
SAGNNOpen-world relationship predictionICTAI2020link
LANLogic attention based neighborhood aggregation for inductive knowledge graph embeddingAAAI2019link
SLANSLAN: similarity-aware aggregation network for embedding out-of-knowledge-graph entitiesNeurocomputing2022link
ARPAttention-based relation prediction of knowledge graph by incorporating graph and context featuresWISE2022link
TransNSOpen knowledge graph representation learning based on neighbors and semantic affinityJournal of Computer Research and Development2019link

⬆️

Multimodal-Enhanced Model

ModelTitleConference/JournalYearPaper
CatEOntological concept structure aware knowledge transfer for inductive knowledge graph embeddingIJCNN2021link
DKRLRepresentation learning of knowledge graphs with entity descriptionsAAAI2016link
OWEAn open-world extension to knowledge graph completion modelsAAAI2019link
WOWEWeighted aggregator for the open-world knowledge graph completionCCIS2020link
Caps-OWKGCaps-OWKG: a capsule network model for open-world knowledge graphInt. J. Mach. Learn. & Cyber.2021link
OWE-MRCExtracting short entity descriptions for open-world extension to knowledge graph completion modelsAdvances in Knowledge Science and Engineering2021link
OWE-RSTRelation specific transformations for open world knowledge graph completionTextGraphs @ ACL2020link
EmReCoEmbeddings based on relation-specific constraints for open world knowledge graph completionApplied Intelligence2023link
ConMaskOpen-world knowledge graph completionAAAI2018link
SDTSDT: an integrated model for open-world knowledge graph reasoningExpert Systems with Applications2020link
Bi-LinkBi-Link: bridging inductive link predictions from text via contrastive learning of transformers and promptsarXiv2022link
RAILDRAILD: towards leveraging relation features for inductive link prediction in knowledge graphsIJCKG2023link
DMoGDecoupling mixture-of-graphs: unseen relational learning for knowledge graph completion by fusing ontology and textual expertsCOLING2022link
BERTRLInductive relation prediction by BERTAAAI2022link
InductivEInductive learning on commonsense knowledge graph completionIJCNN2021link
FITCARLImproving few-shot inductive learning on temporal knowledge graphs using confidence-augmented reinforcement learningMachine Learning and Knowledge Discovery in Databases2023link
TITerTimeTraveler: Reinforcement learning for temporal knowledge graph forecastingEMNLP2021link
MetaTKGRLearning to sample and aggregate: few-shot reasoning over temporal knowledge graphsNeurIPS2022link
FILTFew-shot inductive learning on temporal knowledge graphs using concept-aware informationAKBC2022link
Taxonomy of inductive KGR approaches

IKGR

⬆️

Benchmarks

Datasets for Static KGR Tasks

Dataset#Entities#Relations#Training Triples#Valid Triples#Test TriplesPaper Link
Countries27121,1102424link
Kinship104258,5441,0681,074link
FB1375,04313316,23211,81647,464link
FB1229,73812291,6389,59511,243link
FB15K14,9511,345483,14250,00059,071link
FB15K23714,505237272,11517,53520,466link
FB20K19,9231,452378,07289,04090,143link
FB5M5,385,3221,19219,193,55650,00050,000link
WN1138,58811110,3615,21221,035link
WN1840,94318141,4425,0005,000link
WN18RR40,5591186,8352,9242,924link
YAGO3-10123,143371,079,0404,9784,982link
YAGO37123,18937420,62350,00050,000link
NELL-99575,492200126,1765,0005,000link

Datasets for Dynamic KGR Tasks

Dataset#Entities#RelationsTemporal#Training#Valid#TestPaper Link
GDELT7,691240Timestemp1,033,270238,765305,241link
ICEWS146,738235Timestemp118,76614,85914,756link
ICEWS05-1510,488251Timestemp386,96246,09246,275link
Wikidata12k12,55424Time Interval2,735,685341,961341,961link
YAGO11k10,62310Time Interval161,54019,52320,026link
YAGO15k15,40334Time Interval110,44113,81513,800link

Datasets for Multi-modal KGR Tasks

Dataset#Entities#RelationsModality#Training#Valid#TestPaper Link
FB-IMG-TXT11,7571,231Image+Text285,85034,86329,580link
FB15K237-IMG14,541237Image272,11517,53520,466link
WN9-IMG-TXT6,5559Image+Text11,7411,3191,337link
WN18-IMG40,94318Image141,4425,0005,000link
MKG-Wikipedia15,000169Image34,1964,2744,276link
MKG-YAGO15,00028Image21,3102,6632,665link
TIVA11,85816Video20,0712,0002,000link

Datasets for Few-shot KGR Tasks

Dataset#Entities#Relations#Triples#Training/Valid/Test SplitsPaper Link
NELL-One68,545358181,10951/5/1link
Wiki-One4,868,2448225,859,240133/16/34link
FB15K-One14,541231281,62475/11/33link

Datasets for Inductive KGR Tasks

DatasetVersionTraining/Test Set#Entities#Relations#TriplesPaper Link
FB15K237v1Train2,0001835,226link
Test1,5001462,404
v2Train3,00020312,085
Test2,0001765,092
v3Train4,00021822,394
Test3,0001879,137
v4Train5,00022233,916
Test3,50020414,554
WN18RRv1Train2,74696,678link
Test92291,991
v2Train6,9541018,968
Test2,923104,863
v3Train12,0781132,150
Test5,084117,470
v4Train3,86199,842
Test7,208915,157
NELL-995v1Train10,915145,540link
Test225141,034
v2Train2,5648810,109
Test4,937795,521
v3Train4,64714220,117
Test4,9211229,668
v4Train2,092779,289
Test3,294618,520
Dataset#Entities#Relations#Training Triples#Test TriplesPaper Link
DBPedia50k24,62435132,3886,459link
Wikidata5M4,579,60982220,496,5146,894link

⬆️

Applications

Question Answering

Illustrative examples of the KGR technique applied to QA systems

QA

ModelTitleConference/JournalYearPaper
KBQA SurveyA survey: complex knowledge base question answeringIEEE ICICSE2022link
KEQAKnowledge graph embedding based question answeringACM WSDM2019link
TRL-KEQAQuestion answering over knowledge base embeddings with triples representation learningNeural Information Processing2021link
TransE-QAKnowledge base question answering system based on knowledge graph representation learningACM ICIAI2020link
CAPKGQAComplex question answering over incomplete knowledge graph as n-ary link predictionIEEE IJCNN2022link
EmbedKGQAImproving multi-hop question answering over knowledge graphs using knowledge base embeddingsACL2020link
PKEEQAPath-enhanced multi-relational question answering with knowledge graph embeddingsarXiv2021link
PA-KGQAPath-aware multi-hop question answering over knowledge graph embeddingIEEE ICTAI2022link
HamQAHierarchy-aware multi-hop question answering over knowledge graphsACM Web Conference2023link
BRGNNQuery path generation via bidirectional reasoning for multihop question answering from knowledge basesIEEE TCDS2023link
GRRNImplicit relation inference with deep path extraction for commonsense question answeringNeural Processing Letters2022link
Li et al.Translational relation embeddings for multi-hop knowledge base question answeringWeb Semantics2022link
DSSAGNKnowledge graph multi-hop question answering based on dependent syntactic semantic augmented graph networksElectronics2024link
Jiao et al.A relation embedding assistance networks for multi-hop question answeringACM TALIP2024link
Zhou et al.Marie and BERT – a knowledge graph embedding based question answering system for chemistryACS Omega2023link
CF-KGQACausality-aware enhanced model for multi-hop question answering over knowledge graphsKnowledge-Based Systems2022link
TwiRGCNTwiRGCN: Temporally Weighted Graph Convolution for Question Answering over Temporal Knowledge GraphsEACL2023link
CRONKGQAQuestion answering over temporal knowledge graphsACL-IJCNLP2021link
TempoQRTempoQR: Temporal question reasoning over knowledge graphsAAAI2021link
CTRNAn improving reasoning network for complex question answering over temporal knowledge graphsApplied Intelligence2022link
EXAQTComplex temporal question answering on knowledge graphsACM CIKM2021link
GATQRTemporal knowledge graph question answering models enhanced with GATIEEE BigData2023link
Prog-TQASelf-improvement programming for temporal knowledge graph question answeringLREC-COLING2024link
GenTKGQATwo-stage generative question answering on temporal knowledge graph using large language modelsACL Findings2024link

⬆️

Recommendation

ModelTitleConference/JournalYearPaper
KGCNKnowledge graph convolutional networks for recommender systemsWWW2019link
KGNCF-RRNNeural collaborative recommendation with knowledge graphIEEE ICKG2020link
KGECFKnowledge graph embedding based collaborative filteringIEEE Access2020link
SurveyA review of explainable recommender systems utilizing knowledge graphs and reinforcement learningIEEE Access2024link
PGPRReinforcement knowledge graph reasoning for explainable recommendationSIGIR2019link
CogERCognition-aware knowledge graph reasoning for explainable recommendationWSDM2023link
Hsu et al.Explainable mutual fund recommendation system developed based on knowledge graph embeddingsApplied Intelligence2022link
Lee et al.GCN-based explainable recommendation using a knowledge graph and a language modelIEEE BigData2023link
Markchom et al.Explainable meta-path based recommender systemsACM TORS2023link
Fu et al.Fairness-aware explainable recommendation over knowledge graphsSIGIR2020link
KRRLKnowledge-aware reasoning with self-supervised reinforcement learning for explainable recommendation in MOOCsNeural Computing and Applications2024link
Ryotaro et al.An explainable recommendation framework based on an improved knowledge graph attention network with massive volumes of side informationKnowledge-Based Systems2022link
RippleNetRippleNet: propagating user preferences on the knowledge graph for recommender systemsCIKM2018link
AKUPMAKUPM: attention-enhanced knowledge-aware user preference model for recommendationKDD2019link
RCoLMUnifying task-oriented knowledge graph learning and recommendationIEEE Access2019link
KGATKGAT: knowledge graph attention network for recommendationKDD2019link
IntentGCIntentGC: a scalable graph convolution framework fusing heterogeneous information for recommendationKDD2019link
AKGEHierarchical attentive knowledge graph embedding for personalized recommendationElectronic Commerce Research and Applications2021link
KPRNExplainable reasoning over knowledge graphs for recommendationAAAI2019link

⬆️

Visual Reasoning

Visual Question Answering

ModelTitleConference/JournalYearPaper
FVQAFVQA: fact-based visual question answeringIEEE TPAMI2018link
Wang et al.Explicit knowledge-based reasoning for visual question answeringIJCAI2017link
GraphhopperGraphhopper: multi-hop scene graph reasoning for visual question answeringISWC2021link
Hypergraph TransformerHypergraph transformer: weakly-supervised multi-hop reasoning for knowledge-based visual question answeringACL2022link
CMRLCross-modality multiple relations learning for knowledge-based visual question answeringACM TOMM2024link
KRISPKRISP: Integrating implicit and symbolic knowledge for open-domain knowledge-based VQACVPR2021link
LLM+(KBret+SGret)Find the gap: knowledge base reasoning for visual question answeringarXiv2024link

Cross-Modal Retrieval

ModelTitleConference/JournalYearPaper
KCRKnowledge-aware cross-modal text-image retrieval for remote sensing imagesIEEE TGRS2022link
MMRGMulti-modal relational graph for cross-modal video moment retrievalCVPR2021link
IRGRMultiple instance relation graph reasoning for cross-modal hash retrievalKnowledge-Based Systems2022link

Scene Graph Generation

ModelTitleConference/JournalYearPaper
GB-NetBridging knowledge graphs to generate scene graphsECCV2020link
HiKER-SGGHiKER-SGG: Hierarchical knowledge enhanced robust scene graph generationCVPR2024link
CGRConfigurable graph reasoning for visual relationship detectionTNNLS2022link
COACHERZero-shot scene graph relation prediction through commonsense knowledge integrationECML PKDD2021link

⬆️

Healthcare Domain

ModelTitleConference/JournalYearPaper
Zhu et al.Multimodal reasoning based on knowledge graph embedding for specific diseasesBioinformatics2022link
Chai et al.Diagnosis method of thyroid disease combining knowledge graph and deep learningIEEE Access2020link
SSI-DDISSI-DDI: substructure-substructure interactions for drug-drug interaction predictionBrief. Bioinform.2021link
KGNNKGNN: knowledge graph neural network for drug-drug interaction predictionIJCAI2020link
SMRSMR: medical knowledge graph embedding for safe medicine recommendationBig Data Res.2021link
PharmKGPharmKG: a dedicated knowledge graph benchmark for biomedical data miningBrief. Bioinform.2021link
KG-PredictKG-Predict: a knowledge graph computational framework for drug repurposingJ. Biomed. Inform.2022link

Business Domain

ModelTitleConference/JournalYearPaper
OpenBGConstruction and applications of billion-scale pre-trained multimodal business knowledge graphICDE2023link
Zhang et al.Knowledge graph embedding in e-commerce applications: attentive reasoning, explanations, and transferable rulesInt. Joint Conf. on Knowledge Graphs2021link
Yang et al.Inferring substitutable and complementary products with knowledge-aware path reasoning based on dynamic policy networkKnowledge-Based Syst.2022link
Mitropoulou et al.Anomaly detection in cloud computing using knowledge graph embedding and machine learning mechanismsJ. Grid Comput.2024link
Kosasih et al.Towards knowledge graph reasoning for supply chain risk management using graph neural networksInt. J. Prod. Res.2022link
Yang et al.Research on enterprise risk knowledge graph based on multi-source data fusionNeural Comput. Appl.2022link
Zhang et al.Billion-scale pre-trained e-commerce product knowledge graph modelICDE2021link

Cybersecurity Domain

ModelTitleConference/JournalYearPaper
SikosCybersecurity knowledge graphsKnowl. Inf. Syst.2023link
Ezekia Gilliard et al.Cybersecurity knowledge graph enabled attack chain detection for cyber-physical systemsComputers and Electrical Engineering2023link
Hu et al.Knowledge graph reasoning for cyber attack detectionIET Commun.2024link

Other Domain

ModelTitleConference/JournalYearPaper
Liang et al.Graph path fusion and reinforcement reasoning for recommendation in MOOCsEduc. Inf. Technol.2023link
Zhou et al.Mining tourist preferences and decision support via tourism-oriented knowledge graphInf. Process. Manag.2024link
Gao et al.Hierarchical knowledge graph learning enabled socioeconomic indicator prediction in location-based social networkThe Web Conference (WWW)2023link
Zeng et al.Combining knowledge graph into metro passenger flow prediction: a split-attention relational graph convolutional networkExpert Syst. Appl.2023link
Liu et al.Multi-source knowledge graph reasoning for ocean oil spill detection from satellite SAR imagesInt. J. Appl. Earth Obs. Geoinf.2023link

⬆️

Challenge and Opportunity

Sparse KGR

ModelTitleConference/JournalYearPaper
HoGRNHoGRN: explainable sparse knowledge graph completion via high-order graph reasoning networkIEEE Trans. on Knowledge and Data Engineering2024link
Jia et al.Application of graph neural network and feature information enhancement in relation inference of sparse knowledge graphJournal of Electronic Science and Technology2023link
KRACLKRACL: contrastive learning with graph context modeling for sparse knowledge graph completionThe Web Conference (WWW)2023link
BERT-ConvEEffective use of BERT in graph embeddings for sparse knowledge graph completionACM/SIGAPP Symposium on Applied Computing (SAC)2022link
DacKGRDynamic anticipation and completion for multi-hop reasoning over sparse knowledge graphEMNLP2020link
RuMER-RLRuMER-RL: a hybrid framework for sparse knowledge graph explainable reasoningInformation Sciences2024link
WARWalk-and-relate: a random-walk-based algorithm for representation learning on sparse knowledge graphsarXiv preprint2022link

Uncertain KGR

ModelTitleConference/JournalYearPaper
BEUrREProbabilistic box embeddings for uncertain knowledge graph reasoningNAACL-HLT2021link
SUKESUKE: embedding model for prediction in uncertain knowledge graphIEEE Access2021link
MUKGEEmbedding uncertain knowledge graphsAAAI2019link
UKRMUncertain knowledge graph completion with rule miningWeb Information Systems and Applications2024link
TensorLogTensorlog: a differentiable deductive databasearXiv preprint2016link

KG Error Detection

ModelTitleConference/JournalYearPaper
CKG-EDContrastive knowledge graph error detectionCIKM2022link
CAGEDWhat is normal, what is strange, and what is missing in a knowledge graph: unified characterization via inductive summarizationWWW2020link
HEARKnowledge graph error detection with hierarchical path structureCIKM2023link

Trustworthy KGR

ModelTitleConference/JournalYearPaper
SurveyLogical rule-based knowledge graph reasoning: a comprehensive surveyMathematics2023link
Power-LinkPath-based explanation for knowledge graph completionKDD2024link
IterEIteratively learning embeddings and rules for knowledge graph reasoningWWW2019link
EngineKGPerform like an engine: A closed-loop neural-symbolic learning framework for knowledge graph inferenceCOLING2022link
StreamLearnerLearning temporal rules from knowledge graph streamsAAAI Spring Symposium2019link
TlogicTlogic: temporal logical rules for explainable link forecasting on temporal knowledge graphsAAAI2022link
LCGELogic and commonsense-guided TKGCAAAI2023link
TILPTILP: differentiable learning of temporal logical rules on knowledge graphsICLR2023link
Xu et al.A human-centric evaluation platform for explainable knowledge graph completionEACL (System Demonstrations)2024link
RLF-KGAdvancing Abductive Reasoning in Knowledge Graphs through Complex Logical Hypothesis GenerationACL (Volume 1: Long Papers)2024lihk

LLM-enhanced KGR

ModelTitleConference/JournalYearPaper
KG-GPTKG-GPT: a general framework for reasoning on knowledge graphs using large language modelsFindings of EMNLP2023link
MPIKGCMulti-perspective improvement of knowledge graph completion with large language modelsLREC-COLING2024lihk
LARKComplex logical reasoning over knowledge graphs using large language modelsarXiv2023link
ChatruleChatrule: mining logical rules with large language models for knowledge graph reasoningarXiv2023link
LLM-DALarge language models-guided dynamic adaptation for temporal knowledge graph reasoningarXiv2024link
Xia et al.Chain-of-history reasoning for temporal knowledge graph forecastingFindings of ACL2024link
Luo et al.Chain of history: learning and forecasting with LLMs for temporal knowledge graph completionarXiv2024link
Nguyen et al.Direct evaluation of chain-of-thought in multi-hop reasoning with knowledge graphsFindings of ACL2024link
GenTKGGenTKG: Generative Forecasting on Temporal Knowledge Graph with Large Language ModelsFindings of NAACL2024link

⬆️

awesome
few-shot-learning
inductive-reasoning
knowledge-graph
knowledge-graph-completion
knowledge-graph-embedding
knowledge-graph-reasoning
large-language-model
multi-modal-knowledge-graph
multi-step-reasoning
survey
temporal-reasoning

Similar repos

zjukg/KG-LLM-Papers

[Paper List] Papers integrating knowledge graphs (KGs) and large language models (LLMs)

2,233

XiaoxinHe/Awesome-Graph-LLM

A collection of AWESOME things about Graph-Related LLMs.

2,447

PeterGriffinJin/Awesome-Language-Model-on-Graphs

A curated list of papers and resources based on "Large Language Models on Graphs: A Comprehensive…

996

pengboci/GraphRAG-Survey

290

Graph-RAG/GraphRAG

598

DEEP-PolyU/Awesome-GraphRAG

Awesome-GraphRAG: A curated list of resources (surveys, papers, benchmarks, and opensource…

2,658

woojeongjin/dynamic-KG

Dynamic (Temporal) Knowledge Graph Completion (Reasoning)

610

USTCAGI/Awesome-Papers-Retrieval-Augmented-Generation

89

Contributors

ngl567

34 commits