Imitate, Reward and Introspect: Enhancing Neural Machine Translation by Utilizing Large Language Models as Environments
- 4 November 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Audio, Speech and Language Processing
- Vol. 33, 4736-4747
Abstract
The current neural machine translation (NMT) models face several challenges including the high cost of parallel sentence pairs and difficulties in learning from limited error examples. To address these issues, we propose a novel framework, Large Language Models as Environments (LLMaE), which leverages interactions with large language models (LLMs) to enhance NMT performance. The LLMaE framework comprises three logically sequential steps: behavior imitation, reward feedback, and introspection. The behavior imitation step facilitates the NMT model to mimic the behavioral patterns of the LLMs. The reward feedback step leverages the quality assessments provided by the LLMs to optimize the NMT model and encourage it to produce more accurate translations. Finally, the introspection step enables the NMT model to learn from limited error examples by generating introspective examples. Each stage builds on the previous one, creating a cycle of continuous improvement. Extensive experimental results demonstrate that our proposed method significantly improves the performance of NMT in various scenarios and with different LLMs.Keywords
Funding Information
- National Natural Science Foundation of China (62476271)
- Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS2024316)
This publication has 31 references indexed in Scilit:
- Eliciting the Translation Ability of Large Language Models via Multilingual Finetuning with Translation InstructionsTransactions of the Association for Computational Linguistics, 2024
- Exploring Human-Like Translation Strategy with Large Language ModelsTransactions of the Association for Computational Linguistics, 2024
- Multi-teacher Knowledge Distillation for End-to-End Text Image Machine Translation2023
- Transformer: A General Framework from Machine Translation to OthersMachine Intelligence Research, 2023
- Prompting PaLM for Translation: Assessing Strategies and Performance2023
- Few-shot Learning with Multilingual Generative Language Models2022
- Investigating Catastrophic Forgetting During Continual Training for Neural Machine Translation2020
- COMET: A Neural Framework for MT Evaluation2020
- Learning without ForgettingIEEE Transactions on Pattern Analysis and Machine Intelligence, 2017
- Simple statistical gradient-following algorithms for connectionist reinforcement learningMachine Learning, 1992