CMKL: Modality-Aware Continual Learning for Evolving Biomedical Knowledge Graphs

• URL: http://arxiv.org/abs/2605.10510v1
• Date: Mon, 11 May 2026 13:03:37 GMT
• Title: CMKL: Modality-Aware Continual Learning for Evolving Biomedical Knowledge Graphs
• Abstract summary: We propose a CL framework for biomedical KGs that encodes structure, text, and molecules and protects previously learned knowledge.<n>We evaluate CMKL on a 129K-entity biomedical continual benchmark with 10 tasks.
• Score: 11.909154331954456
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Biomedical knowledge graphs are increasingly large, dynamic, and multimodal, driven by rapid advances in biotechnology such as high-throughput sequencing. Machine learning models can infer previously unobserved biomedical relationships and characterize biomedical entities in these graphs, but existing knowledge graph embedding methods and their continual learning extensions either assume static graph structure or fail to exploit multimodal information under evolving data distributions. They also apply uniform regularization across all model parameters, ignoring that different modalities may exhibit distinct forgetting dynamics as the graph evolves. We propose the Continual Multimodal Knowledge Graph Learner (CMKL), a CL framework for biomedical KGs that natively encodes structure, text, and molecules, fuses them through a Mixture-of-Experts (MoE) router, and protects previously learned knowledge with standard EWC regularization and a K-means-diverse multimodal replay buffer. We evaluate CMKL on a 129K-entity biomedical continual benchmark with 10 tasks. On continual biomedical entity classification, CMKL reaches AP 0.591 versus 0.370 for the strongest structural baseline, a 60% gain that is driven by access to multimodal features and preserved across the sequence with near-zero forgetting (AF 0.008). On continual relationship prediction, CMKL reaches AP $0.062$, matching Naive Sequential and EWC (0.058) within seed noise and outperforming Joint Training (0.047, p=0.045) and LKGE (0.039). A frozen-text ablation reaches AP 0.136, more than double any jointly trained model, yet that signal is unreachable by margin-ranking gradients: the greedy-modality asymmetry lives at the representation level, not the fusion level, and MoE routing manages it by suppressing the unreachable modality without forcing it through a learned bottleneck. Code: github.com/yradwan147/cmkl-neurips2026

Related papers

• MedFlow: Class-Aware Multi-Scale Generation for Medical Time-Series Synthesis [61.211117385840396]
  We propose MedFlow, a class-aware flow matching framework for medical time-series synthesis.<n>MedFlow employs a vector-quantized multi-scale tokenizer to represent medical sequences at complementary temporal resolutions.<n> Token Marginal Guidance, which incorporates class-conditional token statistics directly into the flow matching process, steers generation toward class-specific regions.
  arXiv  Detail & Related papers  (2026-09-04T07:02:12Z)
• BioSync: Transformer-Based Cross-Modal Fusion for a Multimodal Physiological Digital Biomarker [0.0]
  BioSync is a continuous composite digital biomarker defined under the BEST framework.<n>The model applies multi-head self-attention to modality tokens and adds a linear branch whose hypothesis class includes standard feature concatenation.<n>We evaluated BioSync on two literature-informed synthetic cohorts.
  arXiv  Detail & Related papers  (2026-09-03T21:45:18Z)
• MedMamba: Multi-View State Space Models with Adaptive Graph Learning for Medical Time Series Classification [60.41879345743941]
  We propose MedMamba, an end-to-end architecture that integrates state space models with domain-specific inductive biases.<n>MedMamba achieves state-of-the-art performance while maintaining linear computational complexity.
  arXiv  Detail & Related papers  (2026-05-24T09:23:39Z)
• PrimeKG-CL: A Continual Graph Learning Benchmark on Evolving Biomedical Knowledge Graphs [19.859134741964805]
  Existing continual graph learning has been studied almost exclusively on synthetic random, generic KGs.<n>We introduce a benchmark built from nine authoritative biomedical databases.<n>We evaluate six CL strategies across four KGE decoders, plus LKGE, an CK-hugging agent, and CMKL.
  arXiv  Detail & Related papers  (2026-05-11T13:14:02Z)
• RareGraph-Synth: Knowledge-Guided Diffusion Models for Generating Privacy-Preserving Synthetic Patient Trajectories in Ultra-Rare Diseases [0.0]
  We propose a knowledge-guided, continuous-time diffusion framework that generates trajectories for ultra-rare diseases.<n>RareGraph- Synth unifies five public resources into a heterogeneous knowledge graph comprising approximately 8 M typed edges.<n>Timestamped sequences of lab-code, medication-code, and adverse-event-flag triples that contain no protected health information are produced.
  arXiv  Detail & Related papers  (2025-10-06T03:59:09Z)
• impuTMAE: Multi-modal Transformer with Masked Pre-training for Missing Modalities Imputation in Cancer Survival Prediction [75.43342771863837]
  We introduce impuTMAE, a novel transformer-based end-to-end approach with an efficient multimodal pre-training strategy.<n>It learns inter- and intra-modal interactions while simultaneously imputing missing modalities by reconstructing masked patches.<n>Our model is pre-trained on heterogeneous, incomplete data and fine-tuned for glioma survival prediction using TCGA-GBM/LGG and BraTS datasets.
  arXiv  Detail & Related papers  (2025-08-08T10:01:16Z)
• BioGraphFusion: Graph Knowledge Embedding for Biological Completion and Reasoning [0.8983181722105922]
  We introduce BioGraphFusion, a novel framework for deeply synergistic semantic and structural learning.<n> Experiments across three key biomedical tasks demonstrate BioGraphFusion's superior performance over state-of-the-art KE, GNN, and ensemble models.
  arXiv  Detail & Related papers  (2025-07-19T04:03:42Z)
• Multimodal Contrastive Representation Learning in Augmented Biomedical Knowledge Graphs [2.006175707670159]
  PrimeKG++ is an enriched knowledge graph incorporating multimodal data.<n>Our approach demonstrates strong generalizability, enabling accurate link predictions even for unseen nodes.
  arXiv  Detail & Related papers  (2025-01-03T05:29:12Z)
• GTP-4o: Modality-prompted Heterogeneous Graph Learning for Omni-modal Biomedical Representation [68.63955715643974]
  Modality-prompted Heterogeneous Graph for Omnimodal Learning (GTP-4o) We propose an innovative Modality-prompted Heterogeneous Graph for Omnimodal Learning (GTP-4o)
  arXiv  Detail & Related papers  (2024-07-08T01:06:13Z)
• UniCell: Universal Cell Nucleus Classification via Prompt Learning [76.11864242047074]
  We propose a universal cell nucleus classification framework (UniCell) It employs a novel prompt learning mechanism to uniformly predict the corresponding categories of pathological images from different dataset domains. In particular, our framework adopts an end-to-end architecture for nuclei detection and classification, and utilizes flexible prediction heads for adapting various datasets.
  arXiv  Detail & Related papers  (2024-02-20T11:50:27Z)
• Differentiable Agent-based Epidemiology [71.81552021144589]
  We introduce GradABM: a scalable, differentiable design for agent-based modeling that is amenable to gradient-based learning with automatic differentiation. GradABM can quickly simulate million-size populations in few seconds on commodity hardware, integrate with deep neural networks and ingest heterogeneous data sources.
  arXiv  Detail & Related papers  (2022-07-20T07:32:02Z)
• Scientific Language Models for Biomedical Knowledge Base Completion: An Empirical Study [62.376800537374024]
  We study scientific LMs for KG completion, exploring whether we can tap into their latent knowledge to enhance biomedical link prediction. We integrate the LM-based models with KG embedding models, using a router method that learns to assign each input example to either type of model and provides a substantial boost in performance.
  arXiv  Detail & Related papers  (2021-06-17T17:55:33Z)
• A Systematic Approach to Featurization for Cancer Drug Sensitivity Predictions with Deep Learning [49.86828302591469]
  We train >35,000 neural network models, sweeping over common featurization techniques. We found the RNA-seq to be highly redundant and informative even with subsets larger than 128 features.
  arXiv  Detail & Related papers  (2020-04-30T20:42:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.