other
Retrieval pattern that builds and traverses a graph of entities and relationships alongside chunk retrieval.
No evidence-backed relationships are recorded.
KAIST and Microsoft researchers say a map of named entities cut a search agent's input tokens from 206,500 to 88,100 per query on EnterpriseRAG-Bench. Whether that saving holds outside the benchmark depends on what the map costs to build and keep current.
Reality
Her InfoQ presentation argues that knowledge graphs are the foundation of agentic systems. She scopes that claim to her own experience, and the published transcript stops before the definition section.
Reality
Salesforce's Agentforce team argues that a correctly cited answer can still be wrong because the deciding fact lives in another system. That locates the fix in the graph schema itself; resizing chunks won't touch it.
Reality
A principal AI researcher says his team swapped a graph for a tree-structured memory to get faster incremental updates. That makes agent memory a write problem, not a retrieval contest.
Reality
Lookup and aggregation are opposite retrieval problems, and chunk similarity only answers one of them. The second failure is architectural, not a tuning bug.
Reality
Oakley Capital has bought control of Graphwise, the company behind the open-source GraphDB, from its earlier backers. Terms were not disclosed, and neither were licensing commitments.
Reality
Vector search ranks by meaning, so literal tokens like part numbers and ticket IDs fall just outside the top results. The fix is lexical plus dense retrieval, not a model upgrade.
Reality