1 |
ANLIzing the Adversarial Natural Language Inference Dataset
|
|
|
|
In: Proceedings of the Society for Computation in Linguistics (2022)
|
|
BASE
|
|
Show details
|
|
2 |
Automatic Fact-Checking with Document-level Annotations using BERT and Multiple Instance Learning ...
|
|
|
|
BASE
|
|
Show details
|
|
3 |
FastIF: Scalable Influence Functions for Efficient Model Interpretation and Debugging ...
|
|
|
|
BASE
|
|
Show details
|
|
4 |
Open Aspect Target Sentiment Classification with Natural Language Prompts ...
|
|
|
|
BASE
|
|
Show details
|
|
5 |
Does External Knowledge Help Explainable Natural Language Inference? Automatic Evaluation vs. Human Ratings ...
|
|
|
|
BASE
|
|
Show details
|
|
6 |
ContractNLI: A Dataset for Document-level Natural Language Inference for Contracts ...
|
|
|
|
BASE
|
|
Show details
|
|
7 |
IndoNLI: A Natural Language Inference Dataset for Indonesian ...
|
|
|
|
BASE
|
|
Show details
|
|
8 |
Investigating the Effect of Natural Language Explanations on Out-of-Distribution Generalization in Few-shot NLI ...
|
|
|
|
BASE
|
|
Show details
|
|
9 |
Don't Discard All the Biased Instances: Investigating a Core Assumption in Dataset Bias Mitigation Techniques ...
|
|
|
|
BASE
|
|
Show details
|
|
11 |
Generalization in NLI: Ways (Not) To Go Beyond Simple Heuristics ...
|
|
|
|
BASE
|
|
Show details
|
|
12 |
Scheduled Sampling Based on Decoding Steps for Neural Machine Translation ...
|
|
|
|
Abstract:
Anthology paper link: https://aclanthology.org/2021.emnlp-main.264/ Abstract: Scheduled sampling is widely used to mitigate the exposure bias problem for neural machine translation. Its core motivation is to simulate the inference scene during training by replacing ground-truth tokens with predicted tokens, thus bridging the gap between training and inference. However, vanilla scheduled sampling is merely based on training steps and equally treats all decoding steps. Namely, it simulates an inference scene with uniform error rates, which disobeys the real inference scene, where larger decoding steps usually have higher error rates due to error accumulations. To alleviate the above discrepancy, we propose scheduled sampling methods based on decoding steps, increasing the selection chance of predicted tokens with the growth of decoding steps. Consequently, we can more realistically simulate the inference scene during training, thus better bridging the gap between training and inference. Moreover, we ...
|
|
Keyword:
Computational Linguistics; Machine Learning; Machine Learning and Data Mining; Machine translation; Natural Language Inference; Natural Language Processing
|
|
URL: https://dx.doi.org/10.48448/924y-b333 https://underline.io/lecture/37571-scheduled-sampling-based-on-decoding-steps-for-neural-machine-translation
|
|
BASE
|
|
Hide details
|
|
13 |
Pairwise Supervised Contrastive Learning of Sentence Representations ...
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Finding a Balanced Degree of Automation for Summary Evaluation ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
A Multilingual Benchmark for Probing Negation-Awareness with Minimal Pairs ...
|
|
|
|
BASE
|
|
Show details
|
|
16 |
Universal Sentence Representation Learning with Conditional Masked Language Model ...
|
|
|
|
BASE
|
|
Show details
|
|
17 |
BARThez: a Skilled Pretrained French Sequence-to-Sequence Model ...
|
|
|
|
BASE
|
|
Show details
|
|
18 |
Nearest Neighbour Few-Shot Learning for Cross-lingual Classification ...
|
|
|
|
BASE
|
|
Show details
|
|
20 |
Hy-NLI : a Hybrid system for state-of-the-art Natural Language Inference
|
|
|
|
BASE
|
|
Show details
|
|
|
|