KoBE: Knowledge-Based Machine Translation Evaluation
This addresses the need for reliable translation evaluation in scenarios where reference translations are unavailable, offering a practical tool for researchers and practitioners in natural language processing.
The paper tackles the problem of machine translation evaluation without reference translations by grounding entity mentions in source and candidate sentences against a knowledge base and measuring entity recall. It achieves the highest correlation with human judgments on 9 out of 18 language pairs in the WMT19 benchmark and outperforms BLEU on 4 language pairs.
We propose a simple and effective method for machine translation evaluation which does not require reference translations. Our approach is based on (1) grounding the entity mentions found in each source sentence and candidate translation against a large-scale multilingual knowledge base, and (2) measuring the recall of the grounded entities found in the candidate vs. those found in the source. Our approach achieves the highest correlation with human judgements on 9 out of the 18 language pairs from the WMT19 benchmark for evaluation without references, which is the largest number of wins for a single evaluation method on this task. On 4 language pairs, we also achieve higher correlation with human judgements than BLEU. To foster further research, we release a dataset containing 1.8 million grounded entity mentions across 18 language pairs from the WMT19 metrics track data.