用强化学习选少数最优模型，提速翻译集成并提升质量。
Faster Machine Translation Ensembling with Reinforcement Learning and Competitive Correction
- 用RL动态选少量候选模型，避免全模型推理
- 融合块反馈作奖励，训练时优化整体效果
- 引入竞争修正模块，提升最终译文质量
通过集成神经机器翻译（NMT）模型可获得优于单个模型的翻译质量。现有方法通常采用候选选择模块（CSB）和编码器-解码器融合模块（FB），需对所有$ L $个候选模型进行推理，计算开销为$ Ω(L) $。本文提出基于DQN的智能生成策略SmartGen，改进CSB，仅选择固定数量的候选模型，并为每句输入识别最优组合送入融合模块。以往CSB与FB独立训练，导致性能不佳；SmartGen利用FB的反馈作为训练奖励，实现联合优化。此外，针对早期方法中候选模型未经修改直接传入融合模块的问题，本文引入竞争修正模块（CCB）。在英-印、印-英双向翻译任务上进行了充分实验验证。
原文摘要 · Abstract (English)
Ensembling neural machine translation (NMT) models to produce higher-quality translations than the $L$ individual models has been extensively studied. Recent methods typically employ a candidate selection block (CSB) and an encoder-decoder fusion block (FB), requiring inference across \textit{all} candidate models, leading to significant computational overhead, generally $Ω(L)$. This paper introduces \textbf{SmartGen}, a reinforcement learning (RL)-based strategy that improves the CSB by selecting a small, fixed number of candidates and identifying optimal groups to pass to the fusion block for each input sentence. Furthermore, previously, the CSB and FB were trained independently, leading to suboptimal NMT performance. Our DQN-based \textbf{SmartGen} addresses this by using feedback from the FB block as a reward during training. We also resolve a key issue in earlier methods, where candidates were passed to the FB without modification, by introducing a Competitive Correction Block (CCB). Finally, we validate our approach with extensive experiments on English-Hindi translation tasks in both directions.
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。