Agentic SPARQL: Evaluating SPARQL-MCP-powered Intelligent Agents on the Federated KGQA Benchmark
2026-03-10T08:52:25Z•d5fd761f9ca97dbbc8448557bed9da1e8b11daed3a21072a3d4d07aed056b739
Exploration-Space-TheoryHNSWLLM-agentsLLMs-for-data-collection','open-scientific-databases','rerankingMCPMatryoshka-Representation-LearningPAGProjection-Augmented-GraphQwen2SPARQLT-REXapproximate-nearest-neighborarxivdebiasingfederated-KGQAfederated-queryisotonic-layerknowledge-graphlocation-based-recommendationmultilingual-retrievalonline-insertionsrecommendation-systemssemantic-searchtransformertwo-tower
What happened
Collection of recent IR/ML research papers (arXiv) introducing methods across knowledge-graph question answering, production semantic retrieval, recommendation debiasing, formal location-based recommendation theory, basket prediction, ANN search, LLM-powered database construction, reranking, table retrieval, and graph-based retrieval. Key contributions include: SPARQL-MCP integration for agentic federated KGQA and endpoint discovery; a Qwen2 two-tower multilingual retrieval system for Uber Eats with large-scale training and Matryoshka representations; the Isotonic Layer for differentiable, per
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_cs_ir
- Record identifier
- d5fd761f9ca97dbbc8448557bed9da1e8b11daed3a21072a3d4d07aed056b739
- Enrichment time
- 2026-03-10T08:52:25Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.