# Revisiting Round-Trip Translation for Quality Estimation

arXiv: https://arxiv.org/abs/2004.13937 · HTML: https://mlsanity.com/paper/2004.13937/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Apr 2020.

## Built on by (most cited first, 24 of 24)

- [Building Machine Translation Systems for the Next Thousand Languages](https://mlsanity.com/paper/2205.03983/) (2022)
- [IndicXNLI: Evaluating Multilingual Inference for Indian Languages](https://mlsanity.com/paper/2204.08776/) (2022)
- [A Survey on Zero Pronoun Translation](https://mlsanity.com/paper/2305.10196/) (2023)
- [BigVideo: A Large-scale Video Subtitle Translation Dataset for Multimodal Machine Translation](https://mlsanity.com/paper/2305.18326/) (2023)
- [Reference-free Evaluation Metrics for Text Generation: A Survey](https://mlsanity.com/paper/2501.12011/) (2025)
- [Faithful Autoformalization via Roundtrip Verification and Repair](https://mlsanity.com/paper/2604.25031/) (2026)
- [Pretraining Language Models Using Translationese](https://mlsanity.com/paper/2403.13638/) (2024)
- [Improving Gender Translation Accuracy with Filtered Self-Training](https://mlsanity.com/paper/2104.07695/) (2021)
- [Conformalizing Machine Translation Evaluation](https://mlsanity.com/paper/2306.06221/) (2023)
- [Rethinking Round-Trip Translation for Machine Translation Evaluation](https://mlsanity.com/paper/2209.07351/) (2022)
- [vec2text with Round-Trip Translations](https://mlsanity.com/paper/2209.06792/) (2022)
- [Quality Estimation Using Round-trip Translation with Sentence Embeddings](https://mlsanity.com/paper/2111.00554/) (2021)
- [Competency-Aware Neural Machine Translation: Can Machine Translation Know its Own Translation Quality?](https://mlsanity.com/paper/2211.13865/) (2022)
- [Lost In Translation: Generating Adversarial Examples Robust to Round-Trip Translation](https://mlsanity.com/paper/2307.12520/) (2023)
- [Pula: Training Large Language Models for Setswana](https://mlsanity.com/paper/2408.02239/) (2024)
- [Round-Trip Translation Reveals What Frontier Multilingual Benchmarks Miss](https://mlsanity.com/paper/2604.12911/) (2026)
- [Evaluating Inter-Bilingual Semantic Parsing for Indian Languages](https://mlsanity.com/paper/2304.13005/) (2023)
- [BiSync: A Bilingual Editor for Synchronized Monolingual Texts](https://mlsanity.com/paper/2306.00400/) (2023)
- [Parallel Universes, Parallel Languages: A Comprehensive Study on LLM-based Multilingual Counterfactual Example Generation](https://mlsanity.com/paper/2601.00263/) (2026)
- [SCP-NL2TL: Selective Conformal Prediction with Semantic Verification for Natural Language to Temporal Logic Specifications](https://mlsanity.com/paper/2608.05439/) (2026)
- [Applying Automated Machine Translation to Educational Video Courses](https://mlsanity.com/paper/2301.03141/) (2023)
- [RT-SFT: Text Style Transfer from Non-Parallel Corpora by Roundtrip Translation](https://mlsanity.com/paper/2602.15013/) (2026)
- [Is Textual Similarity Invariant under Machine Translation? Evidence Based on the Political Manifesto Corpus](https://mlsanity.com/paper/2605.00618/) (2026)
- [The Language-Energy Divide: Measuring Energy Costs of Multilingual LLM Inference](https://mlsanity.com/paper/2606.21869/) (2026)

## Abstract (arXiv)

Quality estimation (QE) is the task of automatically evaluating the quality of translations without human-translated references. Calculating BLEU between the input sentence and round-trip translation (RTT) was once considered as a metric for QE, however, it was found to be a poor predictor of translation quality. Recently, various pre-trained language models have made breakthroughs in NLP tasks by providing semantically meaningful word and sentence embeddings. In this paper, we employ semantic embeddings to RTT-based QE. Our method achieves the highest correlations with human judgments, compared to previous WMT 2019 quality estimation metric task submissions. While backward translation models can be a drawback when using RTT, we observe that with semantic-level metrics, RTT-based QE is robust to the choice of the backward translation system. Additionally, the proposed method shows consistent performance for both SMT and NMT forward translation systems, implying the method does not penalize a certain type of model.