The paper formalizes implied entailment as a four-way NLI label, builds the INLI dataset from existing implicature sources, and shows fine-tuned models can label implied vs explicit entailment and generalize across conversational and situational domains.
Regularization techniques for fine-tuning in neural machine translation
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
We investigate techniques for supervised domain adaptation for neural machine translation where an existing model trained on a large out-of-domain dataset is adapted to a small in-domain dataset. In this scenario, overfitting is a major challenge. We investigate a number of techniques to reduce overfitting and improve transfer learning, including regularization techniques such as dropout and L2-regularization towards an out-of-domain prior. In addition, we introduce tuneout, a novel regularization technique inspired by dropout. We apply these techniques, alone and in combination, to neural machine translation, obtaining improvements on IWSLT datasets for English->German and English->Russian. We also investigate the amounts of in-domain training data needed for domain adaptation in NMT, and find a logarithmic relationship between the amount of training data and gain in BLEU score.
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Entailed Between the Lines: Incorporating Implication into NLI
The paper formalizes implied entailment as a four-way NLI label, builds the INLI dataset from existing implicature sources, and shows fine-tuned models can label implied vs explicit entailment and generalize across conversational and situational domains.