search

facebookresearch / BLINK

Entity Linker solution
MIT License
1.17k stars 231 forks source link

Benchmark Results #51

Closed mehrdad-naser-73 closed 3 years ago

mehrdad-naser-73 commented 3 years ago

Hi,

I would like to ask a question regarding the benchmark results on widely used EL datasets like AIDA-YAGO and msnbc. I noticed that the reported accuracies are much lower than older non-transformer-based models which is surprising considering the high recall@100. Is this purely due to not fine-tuning on the training set of those datasets?

ledw commented 3 years ago

@mehrdad-naser-73 Hi, thanks for the questions. Yes it is largely due to not fine-tuning on the training set. We have numbers comparable when we fine-tune on the datasets.

mehrdad-naser-73 commented 3 years ago

Thanks for the clarification.