Bidirectional
ثنائي الاتجاه
قدرة النموذج على النظر إلى السياق من الجانبين الأيسر والأيمن في آن واحد عند بناء تمثيل كل رمز.
The model's ability to look at context from both left and right sides simultaneously when building each token's representation.
Also translated asذو اتجاهين، مزدوج الاتجاه
First appears in this corpus in: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (2018)
Appears in these papers
- BART: Denoising Sequence-to-Sequence Pre-Training for Natural Language Generation, Translation, and Comprehension2019in the sky ✦
- BEiT: BERT Pre-Training of Image Transformers2021in the sky ✦
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding2018in the sky ✦
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding2018in the sky ✦
- BLIP-2: Bootstrapping Language-Image Pre-Training with Frozen Image Encoders and Large Language Models2023in the sky ✦
- ELECTRA: Pre-Training Text Encoders as Discriminators Rather Than Generators2020in the sky ✦
- Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions2018in the sky ✦
- XLNet: Generalized Autoregressive Pretraining for Language Understanding2019in the sky ✦
- XLNet: Generalized Autoregressive Pretraining for Language Understanding2019in the sky ✦