# Learning from Many Voices: Literary MT Using Multi-Reference Human and Synthetic Data

arXiv: https://arxiv.org/abs/2412.18707 · HTML: https://mlsanity.com/paper/2412.18707/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in LLM synthetic data generation, one of 13 directions in Synthetic data and self-improvement for LLMs.
- Builds on 22 of its 55 references that are in the map.
- First posted on arXiv 24 Dec 2024.

## Placed in

- Synthetic data and self-improvement for LLMs / [LLM synthetic data generation](https://mlsanity.com/synth/llm-synthetic-data-generation/)

## Builds on (most cited first, 22 of 22)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [mT5: A massively multilingual pre-trained text-to-text transformer](https://mlsanity.com/paper/2010.11934/) (2020)
- [COMET: A Neural Framework for MT Evaluation](https://mlsanity.com/paper/2009.09025/) (2020)
- [Improving Neural Machine Translation Models with Monolingual Data](https://mlsanity.com/paper/1511.06709/) (2015)
- [On LLMs-Driven Synthetic Data Generation, Curation, and Evaluation: A Survey](https://mlsanity.com/paper/2406.15126/) (2024)
- [Marian: Fast Neural Machine Translation in C++](https://mlsanity.com/paper/1804.00344/) (2018)
- [Fine-grained Hallucination Detection and Editing for Language Models](https://mlsanity.com/paper/2401.06855/) (2024)
- [Beyond BLEU: Training Neural Machine Translation with Semantic Similarity](https://mlsanity.com/paper/1909.06694/) (2019)
- [On the Impact of Various Types of Noise on Neural Machine Translation](https://mlsanity.com/paper/1805.12282/) (2018)
- [Democratizing Neural Machine Translation with OPUS-MT](https://mlsanity.com/paper/2212.01936/) (2022)
- [Read before Generate! Faithful Long Form Question Answering with Machine Reading](https://mlsanity.com/paper/2203.00343/) (2022)
- [Exploring Document-Level Literary Machine Translation with Parallel Paragraphs from World Literature](https://mlsanity.com/paper/2210.14250/) (2022)
- [Simple and Effective Paraphrastic Similarity from Parallel Translations](https://mlsanity.com/paper/1909.13872/) (2019)
- [Simulated Multiple Reference Training Improves Low-Resource Machine Translation](https://mlsanity.com/paper/2004.14524/) (2020)
- [ParaGuide: Guided Diffusion Paraphrasers for Plug-and-Play Textual Style Transfer](https://mlsanity.com/paper/2308.15459/) (2023)
- [Multi-Reference Training with Pseudo-References for Neural Translation and Text Generation](https://mlsanity.com/paper/1808.09564/) (2018)
- [Investigating Backtranslation in Neural Machine Translation](https://mlsanity.com/paper/1804.06189/) (2018)
- [Beyond Noise: Mitigating the Impact of Fine-grained Semantic Divergences on Neural Machine Translation](https://mlsanity.com/paper/2105.15087/) (2021)
- [ParaFusion: A Large-Scale LLM-Driven English Paraphrase Dataset Infused with High-Quality Lexical and Syntactic Diversity](https://mlsanity.com/paper/2404.12010/) (2024)
- [Parameter Efficient Diverse Paraphrase Generation Using Sequence-Level Knowledge Distillation](https://mlsanity.com/paper/2404.12596/) (2024)

## Built on by (most cited first, 3 of 3)

- [References Matter: Investigating the Impact of Reference Set Variation on Summarization Evaluation](https://mlsanity.com/paper/2506.14335/) (2025)
- [PaliBench: A Multi-Reference Blueprint for Classical Language Translation Benchmarks](https://mlsanity.com/paper/2605.16881/) (2026)
- [From Outliers to Errors: Auditing Pali-to-English LLM Translations with Multi-Reference Adjudication](https://mlsanity.com/paper/2606.01136/) (2026)

## Abstract (arXiv)

Unlike many other texts, literary works are often translated multiple times. We investigate strategies for leveraging these multi-reference datasets to improve literary machine translation. We propose a filtering framework based on semantic similarity to identify source texts whose references display meaningful variation while remaining faithful. We find that fine-tuning with medium to high semantic similarity data substantially outperforms low semantic similarity data. Moreover, using medium and high semantic similarity data achieves comparable or better performance than using the full unfiltered data. Synthetic translations generated by LLMs are economical and convenient alternatives to human expert translations; however, we find fine-tuning on human expert translations outperforms fine-tuning on synthetically augmented data in automatic metrics and human evaluations, demonstrating the indispensable value of human expert translations for fine-tuning literary machine translation models.