用强化学习选最佳翻译模型，提升非洲低资源语言的领域适配效果。
In-Domain African Languages Translation Using LLMs and Multi-armed Bandits
- 基于多臂老虎机算法动态选择最优翻译模型。
- 在三种非洲语言上验证，无目标数据时仍表现稳健。
- 适合无法微调、数据稀缺的低资源语言场景。
神经机器翻译（NMT）系统在处理低资源语言时面临严峻挑战，尤其在领域适应任务中。受限于训练数据稀少和模型泛化能力不足，选择合适的翻译模型对提升领域内性能至关重要，特别是在无法或不便进行微调的情况下。本文研究了基于带索引算法（如上限置信区间、线性UCB、神经线性带索引、汤普森采样）的模型选择策略。该方法通过高置信度选择实现高效模型筛选，有效缓解资源限制。我们在三种非洲语言及多个领域上评估了该方法，结果表明其在有目标数据和无目标数据两种场景下均具备鲁棒性和有效性。
原文摘要 · Abstract (English)
Neural Machine Translation (NMT) systems face significant challenges when working with low-resource languages, particularly in domain adaptation tasks. These difficulties arise due to limited training data and suboptimal model generalization, As a result, selecting an optimal model for translation is crucial for achieving strong performance on in-domain data, particularly in scenarios where fine-tuning is not feasible or practical. In this paper, we investigate strategies for selecting the most suitable NMT model for a given domain using bandit-based algorithms, including Upper Confidence Bound, Linear UCB, Neural Linear Bandit, and Thompson Sampling. Our method effectively addresses the resource constraints by facilitating optimal model selection with high confidence. We evaluate the approach across three African languages and domains, demonstrating its robustness and effectiveness in both scenarios where target data is available and where it is absent.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。