Content

AI, XR, DevOps, graph theory, knowledge graphs, and digital sovereignty. Deep dives, tutorials, and analysis from across the GraphWiz knowledge base.

All211 AI116 Graphs64 GraphRAG5 Knowledge Graphs6 Visualization1 Algorithms2 Security7 GNN1 LLMs & Graphs2 Agentic AI3

Graphs →AI →

August 12, 2026· ~7 min read

NL2GQL and Text2Cypher in 2026: The New NL2SQL

Natural-language-to-graph-query is the fastest-moving cell in graph research: GQLBench, dialect diversity, agentic tooling, production expectations.

nl2gqltext2cyphergqlgraph-query-languages+4 more

August 12, 2026· ~6 min read

Temporal Knowledge Graphs in 2026: Time-Aware Embeddings and Event-Driven Updates

Temporal KGs are a clear growth cell: time-aware embeddings and event-driven updates as frontiers. Models, trade-offs, production patterns.

temporal-graphsdynamic-graphsknowledge-graphs

← Previous 1...8 9 10...36 Next →

time-aware-embeddings

+3 more

August 7, 2026· ~5 min read

GQL Meets Threat Intelligence: Querying Provenance Graphs for Cyber Attack Investigation

A GQL-style language traces cyber attacks across provenance graphs of audit events — the pattern, a worked example, and what it means for practitioners.

gqlprovenancecyber-securitythreat-intelligence+4 more

August 6, 2026· ~5 min read

Agent Graphs as Memory: When the Agent's Mental Scratchpad Is a Graph

The 2026 agent-memory wave is graph-shaped: persona graphs, experience graphs, self-evolution, and benchmarks proving graph memory beats flat lists.

agentsagent-memorygraph-memoryknowledge-graphs+3 more

August 5, 2026· ~4 min read

GraphRAG Grows Up: Benchmarks, Robustness and the Poisoning-Attack Frontier

GraphRAG left "can it work?" and entered "can you trust it?": benchmarks, robustness to knowledge poisoning, and attacks now hitting graphs.

graphragsecuritybenchmarksrobustness+3 more

August 4, 2026· ~8 min read

Serving Frontier MoE Models at Scale: FP8 KV Caches, INT4 Weights, and KV Cache Integrity

How to squeeze frontier MoE models like Kimi K2.6 and GLM 5.2 onto GPU memory: quantised KV caches double context, INT4 weights cut checkpoint size 40%, and page-level integrity checks protect shared caches. With real SGLang configs and benchmarks.

sglangfp8int4kv-cache+6 more