# Are BLEU and Meaning Representation in Opposition?

arXiv: https://arxiv.org/abs/1805.06536 · HTML: https://mlsanity.com/paper/1805.06536/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 May 2018.

## Built on by (most cited first, 5 of 5)

- [Do Multilingual Neural Machine Translation Models Contain Language Pair Specific Attention Heads?](https://mlsanity.com/paper/2105.14940/) (2021)
- [NMTScore: A Multilingual Analysis of Translation-based Text Similarity Measures](https://mlsanity.com/paper/2204.13692/) (2022)
- [Rationalizing Predictions by Adversarial Information Calibration](https://mlsanity.com/paper/2301.06009/) (2023)
- [On the Role of Pre-trained Language Models in Word Ordering: A Case Study with BART](https://mlsanity.com/paper/2204.07367/) (2022)
- [Intrinsic vs. Extrinsic Evaluation of Czech Sentence Embeddings: Semantic Relevance Doesn't Help with MT Evaluation](https://mlsanity.com/paper/2506.20203/) (2025)

## Abstract (arXiv)

One of possible ways of obtaining continuous-space sentence representations is by training neural machine translation (NMT) systems. The recent attention mechanism however removes the single point in the neural network from which the source sentence representation can be extracted. We propose several variations of the attentive NMT architecture bringing this meeting point back. Empirical evaluation suggests that the better the translation quality, the worse the learned sentence representations serve in a wide range of classification and similarity tasks.