# Knowledge Graph

11 related articles

Subscribe to updates

arXiv Nov 10, 2026
Megrez moves textbook topology formalization into an MK set-theory library: 262 files and 2,529 Qed proofs indexed
For Rocq libraries built over Morse–Kelley set theory, Megrez couples a general coding agent with a rebuildable knowledge graph, a short-burst interactive proving loop and a three-check acceptance pipeline to formalize a textbook chapter by chapter. On the point-set topology corpus MKTopos it covers 12 modules, 262 .v files, 3,894 declarations and 2,529 completed Qed proofs across 102,122 lines of proof script; in retrieval benchmarks, exact-identifier lookup at 100,000 nodes drops from 72.29 ms to 1.23 ms in p95 latency.
For Rocq libraries built over Morse–Kelley set theory, Megrez couples a general coding agent with a rebuildable knowledge graph, a short-burst interactive proving loop and a three-check acceptance pipeline to formalize a textbook chapter by chapter. On the point-set topology corpus MKTopos it covers 12 modules, 262 .v files, 3,894 declarations and 2,529 completed Qed proofs across 102,122 lines of proof script; in retrieval benchmarks, exact-identifier lookup at 100,000 nodes drops from 72.29 ms to 1.23 ms in p95 latency.
Tiancheng Xu（First author）
Megrez Knowledge Graph Formal Theorem Proving Morse–Kelley set theory MKTopos
Graphify-Labs/graphify Oct 16, 2026
Graphify maps a project into a queryable knowledge graph: code parsing stays local with no LLM, and every edge is tagged extracted or inferred
Graphify-Labs' graphify lets an AI coding assistant type /graphify in a project and map code, docs, PDFs, images and video into a queryable knowledge graph, replacing repeated file searching with graph traversal. Code is parsed with tree-sitter into an abstract syntax tree, deterministically, with no LLM and nothing leaving the machine; docs, PDFs, images and video go through a semantic pass using the assistant's model or a configured API key. Every connection is tagged EXTRACTED (explicit in the source) or INFERRED (resolved by graphify), and the output includes a clickable graph.html, GRAPH_REPORT.md and the full graph.json.
Graphify-Labs' graphify lets an AI coding assistant type /graphify in a project and map code, docs, PDFs, images and video into a queryable knowledge graph, replacing repeated file searching with graph traversal. Code is parsed with tree-sitter into an abstract syntax tree, deterministically, with no LLM and nothing leaving the machine; docs, PDFs, images and video go through a semantic pass using the assistant's model or a configured API key. Every connection is tagged EXTRACTED (explicit in the source) or INFERRED (resolved by graphify), and the output includes a clickable graph.html, GRAPH_REPORT.md and the full graph.json.
Graphify Knowledge Graph
NVIDIA Technical Blog Oct 27, 2026
NVIDIA Wires AI-Factory Telemetry Into One Causal Graph: 81 Alerts Become One Incident, Saving About 54 Cluster Hours
NVIDIA has introduced Minerva, an AIOps platform for AI factories that feeds telemetry from GPUs, switches, storage and training jobs into a knowledge graph updated as jobs move, then applies supervised and unsupervised models to catch the gray failures static thresholds miss. In one NVIDIA case, 81 separate events triggered by a switch restart were merged into a single incident, saving about 54 hours of cluster uptime; in another, the platform found one quietly degrading NVLink connection among roughly 400,000 interfaces. These figures come from NVIDIA's internal testing and case descriptions, within its supported scenarios.
NVIDIA has introduced Minerva, an AIOps platform for AI factories that feeds telemetry from GPUs, switches, storage and training jobs into a knowledge graph updated as jobs move, then applies supervised and unsupervised models to catch the gray failures static thresholds miss. In one NVIDIA case, 81 separate events triggered by a switch restart were merged into a single incident, saving about 54 hours of cluster uptime; in another, the platform found one quietly degrading NVLink connection among roughly 400,000 interfaces. These figures come from NVIDIA's internal testing and case descriptions, within its supported scenarios.
NVIDIA
NVIDIA Minerva AIOps Root Cause Analysis Anomaly Detection AI factory
researchsquare.com Oct 23, 2026
Subway fire risk stops relying on hand-picked nodes: a graph network selects 22 from a 666-node knowledge graph, lifting high-casualty probability from 6.58% to 53.2%
Subway fire risk assessment has long depended on experts manually choosing Bayesian network nodes, which is subjective and hard to adapt. This pipeline extracts 215 fire keywords and six risk labels from accident records and fire codes with natural language processing, builds a 666-node knowledge graph, classifies nodes with a three-layer graph convolutional network, and combines confidence weighting with K-means clustering to select 22 representative nodes for the Bayesian network. Under coupled risks such as malicious passenger behavior, smoke exhaust failure and delayed alarm, the high-casualty probability rises from a 6.58% baseline to 53.2%.
Subway fire risk assessment has long depended on experts manually choosing Bayesian network nodes, which is subjective and hard to adapt. This pipeline extracts 215 fire keywords and six risk labels from accident records and fire codes with natural language processing, builds a 666-node knowledge graph, classifies nodes with a three-layer graph convolutional network, and combines confidence weighting with K-means clustering to select 22 representative nodes for the Bayesian network. Under coupled risks such as malicious passenger behavior, smoke exhaust failure and delayed alarm, the high-casualty probability rises from a 6.58% baseline to 53.2%.
Shanghai University
Changhong Wang（First author）
KG-GNN-BN Integrated Reasoning Model Subway Fire Risk Assessment Graph Neural Network Knowledge Graph Bayesian Network
NVIDIA Blog Oct 22, 2026
NVIDIA's Minerva maps GPUs, networks and jobs into one live graph so AI factories fix faults before jobs stall
NVIDIA released Minerva, an AIOps platform for end-to-end observability of AI factories built from large GPU fleets. Previously networking, compute, site reliability and machine learning teams each saw a separate slice of the data center, and alerts arrived disconnected. Minerva pools telemetry from GPUs, SuperNICs, switches, cables, storage, schedulers and workloads into a dynamic knowledge graph, then uses supervised and unsupervised learning to detect anomalies, predict failures and correlate events by topology, collapsing an alert storm into one incident with a likely root cause.
NVIDIA released Minerva, an AIOps platform for end-to-end observability of AI factories built from large GPU fleets. Previously networking, compute, site reliability and machine learning teams each saw a separate slice of the data center, and alerts arrived disconnected. Minerva pools telemetry from GPUs, SuperNICs, switches, cables, storage, schedulers and workloads into a dynamic knowledge graph, then uses supervised and unsupervised learning to detect anomalies, predict failures and correlate events by topology, collapsing an alert storm into one incident with a likely root cause.
NVIDIA
NVIDIA Minerva AIOps AI factory Knowledge Graph
researchsquare.com Oct 18, 2026
Patient-safety reasoning leans on a knowledge graph: removing it drops F1 by 0.047 and worsens all three metrics
A patient-safety reasoning system combines a knowledge graph, retrieval-augmented generation and a LLaMA model to extract safety concepts from MIMIC-III and MIMIC-IV records and produce traceable explanations. The complete system reached an F1 of 0.89 for concept extraction; removing the knowledge graph lowered F1 by 0.047, AUROC by 0.084 and AUPRC by 0.110, all significant. Removing retrieval or LoRA adaptation gave smaller, non-significant drops, and retrieval relevance was confirmed only qualitatively by experts.
A patient-safety reasoning system combines a knowledge graph, retrieval-augmented generation and a LLaMA model to extract safety concepts from MIMIC-III and MIMIC-IV records and produce traceable explanations. The complete system reached an F1 of 0.89 for concept extraction; removing the knowledge graph lowered F1 by 0.047, AUROC by 0.084 and AUPRC by 0.110, all significant. Removing retrieval or LoRA adaptation gave smaller, non-significant drops, and retrieval relevance was confirmed only qualitatively by experts.
Linnaeus University Iran University of Medical Sciences
Sharare Taheri Moghadam（First author）
Patient-Safety Concept Extraction Knowledge Graph Retrieval-Augmented Generation LoRA MIMIC-IV
arXiv Oct 14, 2026
PFAS substitutes get more than a candidate list: three agent stages close the loop from lab stock to manufacturability
For PFAS substitution, a three-stage multi-agent workflow links requirement analysis, candidate generation and manufacturability assessment in a closed loop, raising the overall score from 2.99 to 5.47 with the backbone model held fixed.
Massachusetts Institute of Technology
Tarjei Paule Hage（First author）
MARS PFAS Material Substitution Multi-Agent Reasoning System Retrieval-Augmented Generation Knowledge Graph
arXiv Oct 6, 2026
Dual-process framework cuts red-light-running collisions from 48 to 5 in autonomous driving tests
The work proposes a dual-process-inspired hybrid framework in which a TransFuser++ neural planner generates human-like trajectories in routine driving while a meta-cognitive component, using risk fields shaped by knowledge-graph reasoning, decides when to switch to a model predictive control planner; tested in CARLA on out-of-distribution scenarios of vehicles (including an ambulance) running a red light, it reduced vehicle-to-vehicle collisions from 48 to 5 out of 160 simulations, an 89% reduction.
The work proposes a dual-process-inspired hybrid framework in which a TransFuser++ neural planner generates human-like trajectories in routine driving while a meta-cognitive component, using risk fields shaped by knowledge-graph reasoning, decides when to switch to a model predictive control planner; tested in CARLA on out-of-distribution scenarios of vehicles (including an ambulance) running a red light, it reduced vehicle-to-vehicle collisions from 48 to 5 out of 160 simulations, an 89% reduction.
Simon Janssen（First author）
Hybrid Architecture Knowledge Graph Model Predictive Control Trajectory evaluation End-to-End Autonomous Driving
arXiv Sep 30, 2026
XoG recovers missing reasoning paths via graph embeddings, consistently beating untrained KGQA baselines on incomplete knowledge graphs while cutting LLM token use by up to 33%
The work introduces XoG (eXplore-over-Graph), a framework for multi-hop question answering over incomplete knowledge graphs that uses type-level entity-relation statistics to identify candidate relations and KG embeddings to retrieve plausible missing entities, with the LLM acting as semantic selector and reasoner inside an iterative planning-exploration-reasoning process; on WebQSP, CWQ, and the Wikidata-based BRINK benchmark, XoG remains competitive on complete KGs and consistently outperforms comparable methods without task-specific KGQA training under incompleteness, with gains persisting across multiple LLM backbones and LLM token consumption reduced by up to 33% versus a closely related planning-based approach.
The work introduces XoG (eXplore-over-Graph), a framework for multi-hop question answering over incomplete knowledge graphs that uses type-level entity-relation statistics to identify candidate relations and KG embeddings to retrieve plausible missing entities, with the LLM acting as semantic selector and reasoner inside an iterative planning-exploration-reasoning process; on WebQSP, CWQ, and the Wikidata-based BRINK benchmark, XoG remains competitive on complete KGs and consistently outperforms comparable methods without task-specific KGQA training under incompleteness, with gains persisting across multiple LLM backbones and LLM token consumption reduced by up to 33% versus a closely related planning-based approach.
Ola El Khatib（First author）
XoG (eXplore-over-Graph)Knowledge Graph Question Answering Knowledge Graph Large language models
bioRxiv Sep 18, 2026
Bone science answers stop relying on model memory: give the right passage and one model goes from 42% to 78%
Bone science spans biology, mechanics, materials and clinical medicine, and general models answer fluently without pointing to evidence or keeping an expert's correction. BoneGraph embeds 7,449 full-text documents into 248,629 passage vectors and pairs them with a bone knowledge graph of 1,597 concepts and 1,699 causal relations, exposed as five tabs for search, chat, self-correcting reasoning, imaging and image mechanics. On 50 multiple-choice questions the same model answered 21 without retrieval and 39 when the answer-bearing passage was supplied, while keyword retrieval reached only 26 and did not separate from no retrieval.
Bone science spans biology, mechanics, materials and clinical medicine, and general models answer fluently without pointing to evidence or keeping an expert's correction. BoneGraph embeds 7,449 full-text documents into 248,629 passage vectors and pairs them with a bone knowledge graph of 1,597 concepts and 1,699 causal relations, exposed as five tabs for search, chat, self-correcting reasoning, imaging and image mechanics. On 50 multiple-choice questions the same model answered 21 without retrieval and 39 when the answer-bearing passage was supplied, while keyword retrieval reached only 26 and did not separate from no retrieval.
University of Greenwich The Faraday Institution
Jon Valijonov（First author）Gianluca Tozzi（Corresponding author）
BoneGraph Retrieval-Augmented Generation Self-Correcting Reasoning Knowledge Graph D2IM