mlsanityData to 14 Oct 2026

PrimeKG-CL: A Continual Graph Learning Benchmark on Evolving Biomedical Knowledge Graphs

arXiv 2605.10529 ↗First posted 11 May 2026cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Biomedical knowledge graphs underwrite drug repurposing and clinical decision support, yet the upstream ontologies they depend on update on independent cycles that add millions of edges and deprecate hundreds of thousands more between releases. Yet existing continual graph learning has been studied almost exclusively on synthetic random splits of static, generic KGs, a regime that cannot reproduce the asynchronous, structured evolution real biomedical KGs undergo. To this end, we introduce PrimeKG-CL, a CGL benchmark built from nine authoritative biomedical databases (129K+ nodes, 8.1M+ edges, 10 node types, 30 relation types) with two genuine temporal snapshots (June 2021, July 2023; 5.83M edges added, 889K removed, 7.21M persistent), 10 entity-type-grouped tasks, multimodal node features, and a per-task persistent/added/removed test stratification. On three tasks (biomedical relationship prediction, entity classification, KGQA), we evaluate six CL strategies across four KGE decoders, plus LKGE, an LLM-RAG agent, and CMKL. We find that decoder choice and continual learning strategy interact strongly: no single strategy performs best across all decoders, and mismatched combinations can significantly degrade performance. Moreover, only DistMult exhibits a clear separation between persistent and deprecated knowledge, indicating that standard metrics conflate retention of still-valid facts with failure to forget outdated ones; this effect is absent under RotatE. In addition, multimodal features improve entity-level tasks by up to 60%, and a recent CKGE framework (IncDE) failed to scale to our 5.67M-triple base task across five attempts up to 350GB RAM. Data, pipeline, baselines, and the stratified split are released openly. Dataset:huggingface.co/datasets/yradwan147/PrimeKGCL|Code:github.com/yradwan147/primekg-cl-neurips2026

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 28 of its 40 references that are in the map. First posted on arXiv 11 May 2026.

Builds on

28 of 40 references.

References · 28

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  3. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  4. 2017Modeling Relational Data with Graph Convolutional Networkscited by 1,694
  5. 2020Gradient Surgery for Multi-Task Learningcited by 1,256
  6. 2018Experience Replay for Continual Learningcited by 1,095
  7. 2020Domain-Specific Language Model Pretraining for Biomedical Natural Language Processingcited by 1,009
  8. 2014Embedding Entities and Relations for Learning and Inference in Knowledge Basescited by 932
  9. 2016Complex Embeddings for Simple Link Predictioncited by 902
  10. 2019RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Spacecited by 898
  11. 2023Unifying Large Language Models and Knowledge Graphs: A Roadmapcited by 668
  12. 2019Three scenarios for continual learningcited by 614
16 more
  1. 2018Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interferencecited by 521
  2. 2019Online Continual Learning with Maximally Interfered Retrievalcited by 379
  3. 2022Balanced Multimodal Learning via On-the-fly Gradient Modulationcited by 264
  4. 2018Learning Sequence Encoders for Temporal Knowledge Graph Completioncited by 145
  5. 2022Characterizing and overcoming the greedy nature of learning in multi-modal deep neural networkscited by 95
  6. 2022Multi-modal Contrastive Representation Learning for Entity Alignmentcited by 38
  7. 2021Continual Learning of Knowledge Graph Embeddingscited by 35
  8. 2022Lifelong Embedding Learning and Transfer for Growing Knowledge Graphscited by 28
  9. 2024Towards Continual Knowledge Graph Embedding via Incremental Distillationcited by 17
  10. 2024KG-FIT: Knowledge Graph Fine-Tuning Upon Open-World Knowledgecited by 16
  11. 2022MoSE: Modality Split and Ensemble for Multimodal Knowledge Graph Completioncited by 14
  12. 2024Fast and Continual Knowledge Graph Embedding via Incremental LoRAcited by 14
  13. 2022MMKGR: Multi-hop Multi-modal Knowledge Graph Reasoningcited by 13
  14. 2025Multimodal Contrastive Representation Learning in Augmented Biomedical Knowledge Graphscited by 3
  15. 2026When Modalities Remember: Continual Learning for Multimodal Knowledge Graphscited by 2
  16. 2025Elastic Weight Consolidation for Knowledge Graph Continual Learning: An Empirical Evaluationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026CMKL: Modality-Aware Continual Learning for Evolving Biomedical Knowledge Graphs
  2. 2025BioGraphFusion: Graph Knowledge Embedding for Biological Completion and Reasoning
  3. 2026Injecting Structured Biomedical Knowledge into Language Models: Continual Pretraining vs. GraphRAG
  4. 2026ChronoMedKG: A Temporally-Grounded Biomedical Knowledge Graph and Benchmark for Clinical Reasoning
  5. 2026MolBioKG: Grounding Out-of-Graph Molecules in Biomedical Knowledge Graphs via Multi-Resolution Structural Anchoring
  6. 2025K-Paths: Reasoning over Graph Paths for Drug Repurposing and Drug Interaction Prediction
  7. 2026Large Language Models Meet Biomedical Knowledge Graphs for Mechanistically Grounded Therapeutic Prioritization
  8. 2026Open Biomedical Knowledge Graphs at Scale: Construction, Federation, and AI Agent Access with Samyama Graph Database