- Zero-Shot Information Extraction as a Unified Text-to-Triple Translation, Dawn Song et al.
information extraction - Exploiting Curriculum Learning in Unsupervised Neural Machine Translation,
unsupervised nmt - The Volctrans GLAT System: Non-autoregressive Translation Meets WMT21,
non-autoregressive nmt - Conditional Poisson Stochastic Beam Search,
beam search - FewNLU: Benchmarking State-of-the-Art Methods for Few-Shot Natural Language Understanding,
benchmark - DialogueCSE: Dialogue-based Contrastive Learning of Sentence Embeddings,
sentence embedding - Predicting Attention Sparsity in Transformers,
efficient transformer - Text-based NP Enrichment,
relation extraction - Understanding and Overcoming the Challenges of Efficient Transformer Quantization,
efficient transformer - Data Summarization via Bilevel Optimization,
knowledge tracing - Distiller: A Systematic Study of Model Distillation Methods in Natural Language Processing,
distillation