RuCLIP -- new models and experiments: a technical report

Alex Shonenkov; Andrey Kuznetsov; Denis Dimitrov; Tatyana Shavrina; Daniil Chesakov; Anastasia Maltseva; Alena Fenogenova; Igor Pavlov; Anton Emelyanov; Sergey Markov; Daria Bakshandaeva; Vera Shybaeva; Andrey Chertok

RuCLIP -- new models and experiments: a technical report

Computer Vision and Pattern Recognition 2022-02-23 v1 Artificial Intelligence

Authors: Alex Shonenkov , Andrey Kuznetsov , Denis Dimitrov , Tatyana Shavrina , Daniil Chesakov , Anastasia Maltseva , Alena Fenogenova , Igor Pavlov , Anton Emelyanov , Sergey Markov , Daria Bakshandaeva , Vera Shybaeva , Andrey Chertok

View on arXiv ↗ PDF ↗

Abstract

In the report we propose six new implementations of ruCLIP model trained on our 240M pairs. The accuracy results are compared with original CLIP model with Ru-En translation (OPUS-MT) on 16 datasets from different domains. Our best implementations outperform CLIP + OPUS-MT solution on most of the datasets in few-show and zero-shot tasks. In the report we briefly describe the implementations and concentrate on the conducted experiments. Inference execution time comparison is also presented in the report.

Cite

@article{arxiv.2202.10784,
  title  = {RuCLIP -- new models and experiments: a technical report},
  author = {Alex Shonenkov and Andrey Kuznetsov and Denis Dimitrov and Tatyana Shavrina and Daniil Chesakov and Anastasia Maltseva and Alena Fenogenova and Igor Pavlov and Anton Emelyanov and Sergey Markov and Daria Bakshandaeva and Vera Shybaeva and Andrey Chertok},
  journal= {arXiv preprint arXiv:2202.10784},
  year   = {2022}
}

RuCLIP -- new models and experiments: a technical report

Abstract

Cite

Related papers