| 3 |
On the Limitations of Cross-lingual Encoders as Exposed by Reference-Free Machine Translation Evaluation ...
|
|
|
|
Abstract:
Evaluation of cross-lingual encoders is usually performed either via zero-shot cross-lingual transfer in supervised downstream tasks or via unsupervised cross-lingual textual similarity. In this paper, we concern ourselves with reference-free machine translation (MT) evaluation where we directly compare source texts to (sometimes low-quality) system translations, which represents a natural adversarial setup for multilingual encoders. Reference-free evaluation holds the promise of web-scale comparison of MT systems. We systematically investigate a range of metrics based on state-of-the-art cross-lingual semantic representations obtained with pretrained M-BERT and LASER. We find that they perform poorly as semantic encoders for reference-free MT evaluation and identify their two key limitations, namely, (a) a semantic mismatch between representations of mutual translations and, more prominently, (b) the inability to punish "translationese", i.e., low-quality literal translations. We propose two partial ... : ACL2020 Camera Ready (v3: several small fixes, e.g., Unicode errors) ...
|
|
Keyword:
Computation and Language cs.CL; FOS Computer and information sciences
|
|
URL: https://arxiv.org/abs/2005.01196
https://dx.doi.org/10.48550/arxiv.2005.01196
|
|
BASE
|
|
Hide details
|
|
|
| 4 |
On the limitations of cross-lingual encoders as exposed by reference-free machine translation evaluation
|
|
|
|
BASE
|
|
Show details
|
|
|
| 5 |
On aligning OpenIE extractions with Knowledge Bases: A case study
|
|
|
|
BASE
|
|
Show details
|
|
|
|
|
