Published collection
WikiProfile distinguishes between factual encoding and the recall bottleneck
The benchmark breaks factual errors into encoding, recall, and recognition problems, making it easier to pinpoint models’ factual weaknesses.
- Published entries
- 1
- Sources
- 1
- Date range
- 2026-08-13
- Primary labels
- Research · 论文 · Evaluation Benchmarks · Google Research 知识画像框架与 WikiProfile · Google · OpenAI
Published evidence
Every entry keeps its summary and a path back to the source context.
WikiProfile distinguishes between factual encoding and the recall bottleneck
Google Research proposed a knowledge profile framework, saying frontier LLM factual encoding is nearing saturation but recall remains insufficient; the framework divides facts into five profile types and introduces the WikiProfile benchmark with 2,150 Wikipedia facts.
Original source