BLINKG - Schema-Aligned - Datatype: leaderboard
Assigning the RDF datatype of each literal in the schema-aligned scenario (GTFS-Madrid transit CSV files mapped to the Linked-GTFS ontology, which mirrors the GTFS specification) of BLINKG (Castedo et al., Universidade de Santiago de Compostela, 2026), where an LLM fills in the knowledge-graph construction mappings for given input data and an ontology. F1 in percent against 14 gold mappings, judged by knowledge-graph experts and macro-averaged over three runs at temperature 0.3; higher is better.
Metric: F1 (%; expert-validated mappings for the RDF datatype of each literal; 14 gold task instances; GTFS-Madrid transit CSV files mapped to the LinkedGTFS ontology; macro-average of three runs). Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 90 | #741 |
| 2 | Gemini 2.5 Pro | 90 | #262 |
| 3 | Mixtral 8x22B Instruct | 77 | #2052 |
| 4 | O3 | 77 | #224 |
| 5 | Llama 3.3 70B Instruct | 71 | #1353 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=blinkg-schema-aligned-datatype · How It Works · Data refreshed daily, snapshot 2026-11-12.