Diversity-Aware Literary Machine Translation with Multi-Reward Policy Optimization

Zeynep Yirmibeşoğlu Balal, Tunga Güngör


Abstract
Literary translation is a difficult task that not only requires semantic accuracy but also stylistic richness and lexical diversity. Pretrained and supervised fine-tuned Large Language Models (LLMs) can over-rely on safe vocabulary choices, leading to translations that lack lexical variety. To address this problem, we propose a novel diversity-aware multi-objective Group Relative Policy Optimization (GRPO) framework that pushes the limits of open-source translation quality while increasing lexical diversity. We introduce two diversity-aware reward mechanisms, a Leave-One-Out (LOO) marginal contribution reward and a Self-BLEU penalty, balanced alongside neural quality metrics (COMET), lexical overlap (BLEU), and structural constraints. Through experiments on Turkish-English and German-English using Qwen3-14B, we show that our diversity-aware reinforcement learning approach successfully enhances lexical richness alongside translation quality. Our models achieve state-of-the-art open-source performance in literary translation, bridging the gap with leading commercial systems and demonstrating that policy optimization can effectively steer LLMs toward high-quality, lexically diverse outputs.
Anthology ID:
2026.eamt-1.17
Volume:
Proceedings of the 26th Annual Conference of the European Association for Machine Translation (Volume 1)
Month:
June
Year:
2026
Address:
Tilburg, The Netherlands
Editors:
Dimitar Shterionov, Eva Vanmassenhove, Mirella De Sisto, Fred Blain, Javad Pourmostafa Roshan Sharami, Lisa Lepp, Chiara Manna, Argentina Anna Rescigno, Alina Karakanta, Ayla Rigouts Terryn, Manuel Lardelli, Natalia Resende, Elena Murgolo, Janiça Hackenbuchner, Anna Zaretskaya, Miquel Esplà-Gomis, Thierry Etchegoyhen, Dagmar Gromann, Rachel Bawden, Barry Haddow, Sara Szoc, Mikel Forcada, Helena Moniz
Venue:
EAMT
SIG:
Publisher:
European Association for Machine Translation
Note:
Pages:
231–244
Language:
URL:
https://aclanthology.org/2026.eamt-1.17/
DOI:
Bibkey:
Cite (ACL):
Zeynep Yirmibeşoğlu Balal and Tunga Güngör. 2026. Diversity-Aware Literary Machine Translation with Multi-Reward Policy Optimization. In Proceedings of the 26th Annual Conference of the European Association for Machine Translation (Volume 1), pages 231–244, Tilburg, The Netherlands. European Association for Machine Translation.
Cite (Informal):
Diversity-Aware Literary Machine Translation with Multi-Reward Policy Optimization (Balal & Güngör, EAMT 2026)
Copy Citation:
PDF:
https://aclanthology.org/2026.eamt-1.17.pdf