CLewR: Curriculum Learning with Restarts for Machine Translation Preference Learning
This work addresses a key underexplored aspect in machine translation for researchers and practitioners, though it is incremental as it builds on existing preference optimization methods.
The paper tackles the problem of catastrophic forgetting in machine translation preference learning by introducing CLewR, a curriculum learning strategy with restarts, which consistently improves performance across multiple model families and techniques.
Large language models (LLMs) have demonstrated competitive performance in zero-shot multilingual machine translation (MT). Some follow-up works further improved MT performance via preference optimization, but they leave a key aspect largely underexplored: the order in which data samples are given during training. We address this topic by integrating curriculum learning into various state-of-the-art preference optimization algorithms to boost MT performance. We introduce a novel curriculum learning strategy with restarts (CLewR), which reiterates easy-to-hard curriculum multiple times during training to effectively mitigate the catastrophic forgetting of easy examples. We demonstrate consistent gains across several model families (Gemma2, Qwen2.5, Llama3.1) and preference optimization techniques. We publicly release our code at https://github.com/alexandra-dragomir/CLewR.