REVIEW 8 cited by
Towards Foundation Models for Knowledge Graph Reasoning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Foundation models in language and vision have the ability to run inference on any textual and visual inputs thanks to the transferable representations such as a vocabulary of tokens in language. Knowledge graphs (KGs) have different entity and relation vocabularies that generally do not overlap. The key challenge of designing foundation models on KGs is to learn such transferable representations that enable inference on any graph with arbitrary entity and relation vocabularies. In this work, we make a step towards such foundation models and present ULTRA, an approach for learning universal and transferable graph representations. ULTRA builds relational representations as a function conditioned on their interactions. Such a conditioning strategy allows a pre-trained ULTRA model to inductively generalize to any unseen KG with any relation vocabulary and to be fine-tuned on any graph. Conducting link prediction experiments on 57 different KGs, we find that the zero-shot inductive inference performance of a single pre-trained ULTRA model on unseen graphs of various sizes is often on par or better than strong baselines trained on specific graphs. Fine-tuning further boosts the performance.
Forward citations
Cited by 8 Pith papers
-
Fully Inductive Cardinality Estimation
An encoder-decoder GNN over a factor-graph view of RDF KGs estimates BGP cardinalities on entirely unseen graphs without retraining, cutting median q-error roughly in half versus the best baseline.
-
DualG-MRAG: Decoupling Macro-Reasoning and Micro-Matching for Multimodal Retrieval-Augmented Generation
Decoupling macro topological routing from micro visual matching, plus query-driven GNN path decoding, improves multimodal multi-hop retrieval and QA over strong MM-RAG baselines.
-
InductWave: Inductive Multi-Hop Logical Query Answering on Knowledge Graphs
Combining graph wavelet embeddings with Neural Bellman-Ford message passing reduces the layers needed for inductive logical query answering on large knowledge graphs.
-
Towards Text-free Graph Foundation Models: Rethinking Multi-Domain Graph Contrastive Learning
MDGCL pre-trains graph encoders on multiple domains using same-domain discrimination and a downstream domain-attention mechanism, outperforming existing text-free graph foundation models.
-
Learning Efficient and Generalizable Graph Retriever for Knowledge-Graph Question Answering
RAPL combines LLM-rationalized path labels, line graph transformation, and path-based decoding to improve graph retrieval for KGQA, reporting state-of-the-art results on WebQSP and CWQ.
-
AnomalyGFM: Graph Foundation Model for Zero/Few-shot Anomaly Detection
AnomalyGFM aligns learned normal and abnormal prototypes with node-neighbor residual features, enabling zero-shot and few-shot graph anomaly detection across datasets.
-
ExploreGS: Explorable 3D Scene Reconstruction with Virtual Camera Samplings and Diffusion Priors
Adding information-gain-selected virtual views refined by video diffusion priors to 3D Gaussian Splatting improves arbitrary-view rendering quality.
-
Intellectual Property in Graph-Based Machine Learning as a Service: Attacks and Defenses
A systematic review that organizes graph-ML IP protection into model-level and data-level attacks and defenses, and ships a benchmark library, PyGIP.
Discussion (0). Continue with ORCID to comment.