LegoNN：构建模块化编码器-解码器模型
LegoNN: Building Modular Encoder-Decoder Models
AI总结：
LegoNN 提出一种模块化编码器-解码器构建方法，通过将模块接口锚定到离散词汇表上的边缘分布并引入模态无关编码器，使解码器等组件无需微调即可跨 MT 和 ASR 任务复用，实验表明其匹配或超越基线，并在微调后进一步提升性能。
摘要：
最先进的编码器-解码器模型（例如用于机器翻译（MT）或自动语音识别（ASR）的模型）是作为一个原子单元端到端地构建和训练的。模型的任何组件都不能在没有其他组件的情况下被（重新）使用，这使得无法跨任务共享部分组件，例如高资源的解码器。我们描述了 LegoNN，一种构建编码器-解码器架构的方法，使其部分组件可以应用于其他任务而无需任何微调。为了实现这种可重用性，编码器和解码器模块之间的接口被锚定到预定义离散词汇表上的边缘分布序列。我们提出了两种摄取这些边缘分布的方法；一种是可微分的，允许梯度在整个网络中流动，另一种是梯度隔离的。为了使解码器模块能够在不同源语言的 MT 任务之间以及跨其他任务（如 ASR）移植，我们引入了一种模态无关的编码器，该编码器包含一个长度控制机制，用于动态调整编码器的输出长度，以匹配预训练解码器的预期输入长度范围。我们提出了几个实验来证明 LegoNN 模型的有效性：一个训练好的德语-英语（De-En）MT 任务的语言生成 LegoNN 解码器模块可以在 Europarl 英语 ASR 和罗马尼亚语-英语（Ro-En）MT 任务中无需任何微调地重复使用，匹配或超过基线性能。经过微调后，LegoNN 模型将 Ro-En MT 任务提高了 1.5 个 BLEU 点，并在 Europarl ASR 任务上实现了 12.5% 的相对 WER 降低。为了展示该方法的泛化能力，我们由三个模块组成一个 LegoNN ASR 模型——每个模块都是在三个不同数据集上的不同端到端训练模型中学习到的——实现了 19.5% 的整体 WER 降低。
英文摘要：
State-of-the-art encoder-decoder models (e.g. for machine translation (MT) or automatic speech recognition (ASR)) are constructed and trained end-to-end as an atomic unit. No component of the model can be (re-)used without the others, making it impossible to share parts, e.g. a high resourced decoder, across tasks. We describe LegoNN, a procedure for building encoder-decoder architectures in a way so that its parts can be applied to other tasks without the need for any fine-tuning. To achieve this reusability, the interface between encoder and decoder modules is grounded to a sequence of marginal distributions over a pre-defined discrete vocabulary. We present two approaches for ingesting these marginals; one is differentiable, allowing the flow of gradients across the entire network, and the other is gradient-isolating. To enable the portability of decoder modules between MT tasks for different source languages and across other tasks like ASR, we introduce a modality agnostic encoder which consists of a length control mechanism to dynamically adapt encoders' output lengths in order to match the expected input length range of pre-trained decoders. We present several experiments to demonstrate the effectiveness of LegoNN models: a trained language generation LegoNN decoder module from German-English (De-En) MT task can be reused without any fine-tuning for the Europarl English ASR and the Romanian-English (Ro-En) MT tasks, matching or beating the performance of baseline. After fine-tuning, LegoNN models improve the Ro-En MT task by 1.5 BLEU points and achieve 12.5% relative WER reduction on the Europarl ASR task. To show how the approach generalizes, we compose a LegoNN ASR model from three modules -- each has been learned within different end-to-end trained models on three different datasets -- achieving an overall WER reduction of 19.5%.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
