Fusion-in-Decoder (FiD) is a powerful retrieval-augmented language model that sets the state-of-the-art on many knowledge-intensive NLP tasks. However, FiD suffers from very expensive inference. We show that the majority of inference time results from memory bandwidth constraints in the decoder, and propose two simple changes to the FiD architecture to speed up inference by 7x. The faster decoder inference then allows for a much larger decoder. We denote FiD with the above modifications as FiDO, and show that it strongly improves performance over existing FiD models for a wide range of inference budgets. For example, FiDO-Large-XXL performs faster inference than FiD-Base and achieves better performance than FiD-Large.
translated by 谷歌翻译
检索演示的生成模型比独立语言模型提供了许多好处:除了对给定查询的文字答案外,它们还提供了从可更新知识库中检索到的出处项目。但是,它们也是更复杂的系统,需要处理长输入。在这项工作中,我们介绍了FID Light,以强烈提高最先进的检索功能模型的效率,同时保持相同的有效性。我们的FID光模型将信息流从编码器(分别编码段落)限制为解码器(使用串联编码表示)。此外,我们通过文本源指针通过重新排列的功能调整FID光,以提高排名最高的出处精度。我们对七个知识密集任务(KILT)的各种实验表明,FID光线始终改善了查询潜伏期和有效性之间的帕累托前沿。带有源指向的FID光设置为六个苏格兰短裙任务的新最新结果,用于合并文本生成和出处检索评估,同时保持合理的效率。
translated by 谷歌翻译
信息检索的任务是许多自然语言处理系统的重要组成部分,例如开放式域问题回答。尽管传统方法是基于手工制作的功能,但基于神经网络的连续表示最近获得了竞争结果。使用此类方法的一个挑战是获取监督数据以训练回猎犬模型,该模型对应于一对查询和支持文档。在本文中,我们提出了一种技术,以学习以知识蒸馏的启发,并不需要带注释的查询和文档对。我们的方法利用读者模型的注意分数,用于根据检索文档解决任务,以获取猎犬的合成标签。我们评估我们的方法回答,获得最新结果。
translated by 谷歌翻译
We study the problem of efficient generative inference for Transformer models, in one of its most challenging settings: large deep models, with tight latency targets and long sequence lengths. Better understanding of the engineering tradeoffs for inference for large Transformer-based models is important as use cases of these models are growing rapidly throughout application areas. We develop a simple analytical model for inference efficiency to select the best multi-dimensional partitioning techniques optimized for TPU v4 slices based on the application requirements. We combine these with a suite of low-level optimizations to achieve a new Pareto frontier on the latency and model FLOPS utilization (MFU) tradeoffs on 500B+ parameter models that outperforms the FasterTransformer suite of benchmarks. We further show that with appropriate partitioning, the lower memory requirements of multiquery attention (i.e. multiple query heads share single key/value head) enables scaling up to 32x larger context lengths. Finally, we achieve a low-batch-size latency of 29ms per token during generation (using int8 weight quantization) and a 76% MFU during large-batch-size processing of input tokens, while supporting a long 2048-token context length on the PaLM 540B parameter model.
translated by 谷歌翻译
在深度学习中,模型通常重用所有输入的相同参数。专家的混合(MOE)违反了这一点,而是为每个传入示例选择不同的参数。结果是一个稀疏激活的模型 - 具有残酷数量的参数 - 但恒定的计算成本。然而,尽管MOE取得了一些显着的成功,但复杂性,沟通成本和培训不稳定的阻碍了广泛的采用 - 我们使用Switch Transformer解决了这些领域。我们简化了MOE路由算法和设计直观的改进模型,以降低的通信和计算成本。我们提出的培训技术有助于纠缠不稳定,我们表明稀疏模型可能首次以较低的精度(BFLOAT16)格式进行培训。我们设计了基于T5基数和T5总数的模型,以使用相同的计算资源获得高达7倍的训练速度。这些改进扩展到多语言设置,我们在所有101种语言中衡量对MT5基本版本的收益。最后,我们通过在“巨大的清洁爬行语料库”上预先培训高达数万亿个参数模型,并在T5-XXL模型上实现4倍的速度,从而提高了语言模型的当前规模。
translated by 谷歌翻译
最近的开放式域问题的作品应答使用检索器模型引用外部知识库,可选地重新映射与单独的重新编制模型,并使用另一个读取器模型生成答案。尽管执行相关任务,但模型具有单独的参数,并且在训练期间略微耦合。在这项工作中,我们建议将猎犬和重新划分为依次应用于变压器架构内的硬注视机制,并将所产生的计算表示给读者送入。在这个奇异模型架构中,隐藏的表示从搬运者逐渐改进到Reranker到读者,这更有效地利用模型容量,并且当我们以端到端的方式训练时,还导致更好的梯度流动。我们还提出了一种预先训练的方法,以有效地培训这种架构。我们评估我们的自然问题和TriviaQA Open DataSets的模型以及固定参数预算,我们的模型优于以前的最先进模型1.0和0.7精确匹配分数。
translated by 谷歌翻译
大型语言模型在各种任务上显示出令人印象深刻的几次结果。但是,当知识是此类结果的关键时,就像问题回答和事实检查之类的任务一样,似乎需要存储知识的大量参数计数。众所周知,检索增强模型可以在不需要多个参数的情况下在知识密集的任务上表现出色,但是目前尚不清楚它们是否在几个弹药设置中工作。在这项工作中,我们介绍了地图集,这是一个经过精心设计和预先训练的增强语言模型,能够通过很少的培训示例学习知识密集型任务。我们对包括MMLU,苏格兰短裙和归类等各种任务进行评估,并研究文档索引内容的影响,表明它可以很容易地进行更新。值得注意的是,在自然问题上仅使用64个示例在自然问题上达到超过42 \%的准确性,尽管参数少了50倍,但比540B参数模型的表现优于540b参数模型。
translated by 谷歌翻译
从有限的资源中获得最大收益可以进步自然语言处理(NLP)研究和实践,同时保守资源。这些资源可能是数据,时间,存储或能源。NLP的最新工作从缩放率产生了有趣的结果。但是,仅使用比例来改善结果意味着资源消耗也会扩展。这种关系激发了对有效方法的研究,这些方法需要更少的资源才能获得相似的结果。这项调查涉及NLP效率的方法和发现,旨在指导该领域的新研究人员并激发新方法的发展。
translated by 谷歌翻译
Systems for knowledge-intensive tasks such as open-domain question answering (QA) usually consist of two stages: efficient retrieval of relevant documents from a large corpus and detailed reading of the selected documents to generate answers. Retrievers and readers are usually modeled separately, which necessitates a cumbersome implementation and is hard to train and adapt in an end-to-end fashion. In this paper, we revisit this design and eschew the separate architecture and training in favor of a single Transformer that performs Retrieval as Attention (ReAtt), and end-to-end training solely based on supervision from the end QA task. We demonstrate for the first time that a single model trained end-to-end can achieve both competitive retrieval and QA performance, matching or slightly outperforming state-of-the-art separately trained retrievers and readers. Moreover, end-to-end adaptation significantly boosts its performance on out-of-domain datasets in both supervised and unsupervised settings, making our model a simple and adaptable solution for knowledge-intensive tasks. Code and models are available at https://github.com/jzbjyb/ReAtt.
translated by 谷歌翻译
尽管经过验证的大型变压器模型已被证明具有很高的能力解决自然语言任务,但处理长序列输入仍然是一个重大挑战。这样的任务之一就是长输入摘要,其中输入比大多数预验证的模型的最大输入上下文更长。通过一系列广泛的实验,我们研究了哪些模型架构变化和预处理范式可以最有效地适应经过预定的变压器以进行长输入摘要。我们发现,带有全局编码器代币的交错,块状变压器可以达到良好的性能和效率平衡,并且在长序列上有意义地改善了下游摘要性能。根据我们的发现,我们介绍了Pegasus-X,这是Pegasus模型的扩展,并具有额外的长输入预处理,以处理最多16K令牌的输入。 Pegasus-X在长输入摘要任务上实现了强劲的性能,与更大的模型相当,同时添加了很少的其他参数,并且不需要模型并行训练。
translated by 谷歌翻译
我们通过与与前面令牌的局部相似度,通过调节从大语料库检索的文档块来增强自动回归语言模型。尽管使用25美元\时分,我们的检索增强型变压器(RetroCro)的检索增强型变压器(RetroCr)对GPT-3和侏罗纪-1获得了可比性的性能。微调后,复古表演转换为下游知识密集型任务,如问题应答。复古结合了冷冻BERT猎犬,一种可微分的编码器和块状的横向机制,以预测基于数量级的令牌,而不是训练期间通常消耗的数量。我们通常从头开始训练复古,还可以快速改造预先接受的变压器,通过检索,仍然达到良好的性能。我们的工作通过以前所未有的规模开辟了通过显式内存改进语言模型的新途径。
translated by 谷歌翻译
Training large, deep neural networks to convergence can be prohibitively expensive. As a result, often only a small selection of popular, dense models are reused across different contexts and tasks. Increasingly, sparsely activated models, which seek to decouple model size from computation costs, are becoming an attractive alternative to dense models. Although more efficient in terms of quality and computation cost, sparse models remain data-hungry and costly to train from scratch in the large scale regime. In this work, we propose sparse upcycling -- a simple way to reuse sunk training costs by initializing a sparsely activated Mixture-of-Experts model from a dense checkpoint. We show that sparsely upcycled T5 Base, Large, and XL language models and Vision Transformer Base and Large models, respectively, significantly outperform their dense counterparts on SuperGLUE and ImageNet, using only ~50% of the initial dense pretraining sunk cost. The upcycled models also outperform sparse models trained from scratch on 100% of the initial dense pretraining computation budget.
translated by 谷歌翻译
We introduce EdgeFormer -- a parameter-efficient Transformer for on-device seq2seq generation under the strict computation and memory constraints. Compared with the previous parameter-efficient Transformers, EdgeFormer applies two novel principles for cost-effective parameterization, allowing it to perform better given the same parameter budget; moreover, EdgeFormer is further enhanced by layer adaptation innovation that is proposed for improving the network with shared layers. Extensive experiments show EdgeFormer can effectively outperform previous parameter-efficient Transformer baselines and achieve competitive results under both the computation and memory constraints. Given the promising results, we release EdgeLM -- the pretrained version of EdgeFormer, which is the first publicly available pretrained on-device seq2seq model that can be easily fine-tuned for seq2seq tasks with strong results, facilitating on-device seq2seq generation in practice.
translated by 谷歌翻译
大型神经模型的培训和推断很昂贵。但是,对于许多应用程序域,虽然新任务和模型经常出现,但建模的基础文档主要保持不变。我们研究如何通过嵌入回收利用(ER)来降低此类设置的计算成本:在执行训练或推理时从以前的模型中重新使用激活。与以前的工作相反,重点是冻结小型分类头进行填充,这通常会导致绩效显着下降,我们提出了从预告片的模型中缓存中间层的输出,并为新任务的剩余层进行填充。我们表明,我们的方法在训练过程中提供了100%的速度和55-86%的推理,并且对科学领域中文本分类和实体识别任务的准确性产生了可观的影响。对于通用域的问答任务,ER提供了类似的加速和少量准确性。最后,我们确定了ER的几个开放挑战和未来的方向。
translated by 谷歌翻译
我们提出了一种用于在生成答案时将信息与多个检索文件中的信息组合的可检索增强的开放式开放式开放式开放域问题训练方法。我们将检索决策模拟作为相关文件集的潜在变量。由于通过对所检索的文件集的边缘化,因此使用期望最大化算法估计这一点。我们迭代地估计我们的潜在变量的价值(给定问题的这些相关文档集),然后使用此估计来更新检索器和读取器参数。我们假设这种端到端的训练允许训练信号流到读者,然后比上演明智的训练更好地流到猎犬。这导致检索器能够为问题和读者选择更多相关文档,这些文件在更准确的文档中培训以生成答案。三个基准数据集的实验表明,我们所提出的方法优于所有现有的相当大小的方法2-3%绝对精确匹配点,实现了新的最先进的结果。我们的结果还展示了学习检索以改善答复的可行性,而无明确监督检索决策。
translated by 谷歌翻译
已经表明,在一个域上训练的双编码器经常概括到其他域以获取检索任务。一种广泛的信念是,一个双编码器的瓶颈层,其中最终得分仅仅是查询向量和通道向量之间的点产品,它过于局限,使得双编码器是用于域外概括的有效检索模型。在本文中,我们通过缩放双编码器模型的大小{\ em同时保持固定的瓶颈嵌入尺寸固定的瓶颈的大小来挑战这一信念。令人惊讶的是,令人惊讶的是,缩放模型尺寸会对各种缩放提高检索任务,特别是对于域外泛化。实验结果表明,我们的双编码器,\ textbf {g} enovalizable \ textbf {t} eTrievers(gtr),优先级%colbert〜\ cite {khattab2020colbertt}和现有的稀疏和密集的索取Beir DataSet〜\ Cite {Thakur2021Beir}显着显着。最令人惊讶的是,我们的消融研究发现,GTR是非常数据的高效,因为它只需要10 \%MARCO监督数据,以实现最佳域的性能。所有GTR模型都在https://tfhub.dev/google/collections/gtr/1发布。
translated by 谷歌翻译
稀疏的专家模型是一个三十年来的概念,作为深度学习中流行的建筑。这类体系结构包括专家的混合物,交换变压器,路由网络,基础层等,所有这些都以一个统一的想法,即每个示例都由参数的一个子集进行。通过这样做,稀疏度将参数计数与每个示例的计算分解,从而允许使用极大但有效的模型。最终的模型显示了各种领域的显着改善,例如自然语言处理,计算机视觉和语音识别。我们回顾了稀疏专家模型的概念,提供了对常见算法的基本描述,将深度学习时代的进步进行上下文化,并通过突出未来工作的领域来结束。
translated by 谷歌翻译
The pre-training of masked language models (MLMs) consumes massive computation to achieve good results on downstream NLP tasks, resulting in a large carbon footprint. In the vanilla MLM, the virtual tokens, [MASK]s, act as placeholders and gather the contextualized information from unmasked tokens to restore the corrupted information. It raises the question of whether we can append [MASK]s at a later layer, to reduce the sequence length for earlier layers and make the pre-training more efficient. We show: (1) [MASK]s can indeed be appended at a later layer, being disentangled from the word embedding; (2) The gathering of contextualized information from unmasked tokens can be conducted with a few layers. By further increasing the masking rate from 15% to 50%, we can pre-train RoBERTa-base and RoBERTa-large from scratch with only 78% and 68% of the original computational budget without any degradation on the GLUE benchmark. When pre-training with the original budget, our method outperforms RoBERTa for 6 out of 8 GLUE tasks, on average by 0.4%.
translated by 谷歌翻译
我们介绍了Art,这是一种新的语料库级自动编码方法,用于培训密集检索模型,不需要任何标记的培训数据。密集的检索是开放域任务(例如Open QA)的核心挑战,在该任务中,最先进的方法通常需要大量的监督数据集,并具有自定义的硬性采矿和肯定式示例。相反,艺术品仅需要访问未配对的投入和输出(例如问题和潜在的答案文件)。它使用新的文档 - 重新定义自动编码方案,其中(1)输入问题用于检索一组证据文档,并且(2)随后使用文档来计算重建原始问题的概率。基于问题重建的检索培训可以有效地学习文档和问题编码器,以后可以将其纳入完整的QA系统中,而无需任何进一步的填充。广泛的实验表明,ART在多个QA检索基准测试基准上获得最先进的结果,并且仅来自预训练的语言模型的一般初始化,从而消除了对标记的数据和特定于任务的损失的需求。
translated by 谷歌翻译
In this paper, we propose an end-to-end Retrieval-Augmented Visual Language Model (REVEAL) that learns to encode world knowledge into a large-scale memory, and to retrieve from it to answer knowledge-intensive queries. REVEAL consists of four key components: the memory, the encoder, the retriever and the generator. The large-scale memory encodes various sources of multimodal world knowledge (e.g. image-text pairs, question answering pairs, knowledge graph triplets, etc) via a unified encoder. The retriever finds the most relevant knowledge entries in the memory, and the generator fuses the retrieved knowledge with the input query to produce the output. A key novelty in our approach is that the memory, encoder, retriever and generator are all pre-trained end-to-end on a massive amount of data. Furthermore, our approach can use a diverse set of multimodal knowledge sources, which is shown to result in significant gains. We show that REVEAL achieves state-of-the-art results on visual question answering and image captioning.
translated by 谷歌翻译