知识图谱论文趋势 2023–2026

主要结论

• 知识图谱论文从 2023 年的 287 篇增长到 2026 年的 536 篇，三年间扩大到约 1.9 倍。
• ICLR 是发知识图谱论文最多的会议，2023–2026 年共 571 篇，占四个会议总量的 33%。
• 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 17 篇增至 2026 年的 93 篇。
• 在「图与结构化数据」的 5 个主题里，它的 2026 年论文数排第 2；同组增长最快的时间序列是 3.2 倍，知识图谱是 1.9 倍。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• NeurIPS 2026Aligning LLMs with Biomedical Knowledge using Balanced Fine-Tuning
• ICML 2026Belief Propagation Converges to Gaussian Distributions in Sparsely-Connected Factor Graphs
• ICLR 2026$\ell_1$ Latent Distance based Continuous-time Graph Representation
• CVPR 2026A Unified Framework for Knowledge Transfer in Bidirectional Model Scaling
• NeurIPS 2025A CLT for Polynomial GNNs on Community-Based Graphs
• ICML 2025A Selective Learning Method for Temporal Graph Continual Learning
• ICLR 2025$R^2$-Guard: Robust Reasoning Enabled LLM Guardrail via Knowledge-Enhanced Logical Reasoning
• CVPR 2025A Hubness Perspective on Representation Learning for Graph-Based Multi-View Clustering
• NeurIPS 2024Can Large Language Models Analyze Graphs like Professionals? A Benchmark, Datasets and Models
• ICML 2024Bipartite Matching in Massive Graphs: A Tight Analysis of EDCS
• ICLR 2024A Good Learner can Teach Better: Teacher-Student Collaborative Knowledge Distillation
• CVPR 2024De-confounded Data-free Knowledge Distillation for Handling Distribution Shifts
• NeurIPS 2023A Comprehensive Study on Text-attributed Graphs: Benchmarking and Rethinking
• ICML 2023A Generalization of ViT/MLP-Mixer to Graphs
• ICLR 2023A Differential Geometric View and Explainability of GNN on Evolving Graphs
• CVPR 2023A Unified Knowledge Distillation Framework for Deep Directed Graphical Models
• NeurIPS 2026AT-SKM-Net: An Accelerated Trainable Sampling Kaczmarz-Motzkin Framework for Linear Hard-Constraint Feasibility on Dynamic Graphs
• ICML 2026Cross-Modal Knowledge Distillation without Paired Data: Theoretical Foundation and Algorithm
• ICLR 2026A Graph Meta-Network for Learning on Kolmogorov–Arnold Networks
• CVPR 2026Distilling Balanced Knowledge from a Biased Teacher

全部知识图谱论文

共 133 篇，第 3 / 3 页

101. NeurIPS 2026Reviving Stale Updates: Data-Free Knowledge Distillation for Asynchronous Federated Learning
102. NeurIPS 2026RewardFlow: Topology-Aware Reward Propagation on State Graphs for Agentic RL with LLMs
103. NeurIPS 2026Robust and Hard-to-Remove GNN Watermarking via Topological Invariant Perception
104. NeurIPS 2026S3Former: Sequential, Structural, and Statistical Fusion for Continuous-Time Dynamic Graphs
105. NeurIPS 2026SGNNBench: A Holistic Evaluation of Spiking Graph Neural Networks on Large-scale Graphs
106. NeurIPS 2026Similarity-Constrained Reweighting for Complex Query Answering on Knowledge Graphs
107. NeurIPS 2026Simple yet Effective Semi-supervised Knowledge Distillation from Vision-Language Models via Dual-Head Optimization
108. NeurIPS 2026SKIM: Pruning Large Language Model Agents via Selective Knowledge Informed Masking
109. NeurIPS 2026Spatial Representation Distillation and Knowledge Routing for Vision-Language-Action Models
110. NeurIPS 2026Speculative Self-Distillation enables Efficient Knowledge Internalization
111. NeurIPS 2026Stabilizing Knowledge, Promoting Reasoning: Dual-Token Constraints for RLVR
112. NeurIPS 2026The Fractured Interlingua: Geometric Bottlenecks in Cross-Lingual Knowledge Editing
113. NeurIPS 2026The Geometry of Forgetting: Temporal Knowledge Drift as an Independent Axis in LLM Representations
114. NeurIPS 2026The Graph Concept Bottleneck: Decoding Combinatorial Reasoning in GNNs for Interpretability
115. NeurIPS 2026The Illusion of Forgetting: Rank Leakage in Knowledge Editing and Its Mitigation
116. NeurIPS 2026The Override Gap: A Magnitude Account of Knowledge Conflict Failure in Hypernetwork-Based Instant LLM Adaptation
117. NeurIPS 2026The Query Complexity of Local Search in Rounds on General Graphs
118. NeurIPS 2026They Can See It but not Say It: Iterative Self Knowledge Re-expression in Visual Reasoning Relative Pose Identification
119. NeurIPS 2026Towards Explainable Industrial Anomaly Detection via Knowledge-Guided Latent Reasoning
120. NeurIPS 2026Towards Generalist Graph-Level Anomaly Detection
121. NeurIPS 2026Towards Precise Knowledge Distillation for Large Language Models via Knowledge Probing
122. NeurIPS 2026Transformers Provably Learn Graph Search: Training Dynamics and the Exponential Power of Depth
123. NeurIPS 2026Traversal-Invariant Positional Encoding for Serialized Graphs
124. NeurIPS 2026Tree Search With Predictions
125. NeurIPS 2026UFO: A Unified Flow-Oriented Framework for Robust Continual Graph Learning
126. NeurIPS 2026Uncovering Semantic Hierarchies in Text-Attributed Graphs via Variational EM-based LLM–GNN Synergy
127. NeurIPS 2026VOID: Backdoor Injection through Knowledge Vacuity in Federated Unlearning
128. NeurIPS 2026Weisfeiler and Leman Follow the Arrow of Time: Expressive Power of Message Passing in Temporal Event Graphs
129. NeurIPS 2026Weisfeiler-Leman Is Incomplete on Simple Spectrum Graphs, so Canonicalize Them
130. NeurIPS 2026When Does Fine-Tuning Extract Stored Knowledge? A Relation-Covering Theory for One-Layer Transformers
131. NeurIPS 2026xVGAE: A Hierarchical Variational Graph Autoencoder for Exchangeable Graphs
132. NeurIPS 2026You Don’t Need Aligned Representations: Knowledge Distillation via Random Prototype Spaces
133. NeurIPS 2026Your Embedding Model Is SMARTer Than You Think

常见问题

2023–2026 年四大会议一共有多少篇知识图谱论文？

共 1,715 篇，按年份分别是 2023 年 287 篇、2024 年 426 篇、2025 年 466 篇、2026 年 536 篇。

NeurIPS 2026 有多少篇知识图谱论文？

NeurIPS 2026 有 133 篇知识图谱论文，上一届是 154 篇。

ICLR 2026 有多少篇知识图谱论文？

ICLR 2026 有 201 篇知识图谱论文，上一届是 172 篇。

ICML 2026 有多少篇知识图谱论文？

ICML 2026 有 149 篇知识图谱论文，上一届是 99 篇。

CVPR 2026 有多少篇知识图谱论文？

CVPR 2026 有 53 篇知识图谱论文，上一届是 41 篇。

哪个会议的知识图谱论文最多？

ICLR，2023–2026 年共 571 篇。