训练视觉和语言模型的更多数据总是更好吗？我们研究多模式任务中的知识可传递性。当前的机器学习趋势是假设通过从不同任务加入多个数据集，其整体绩效将有所改善。但是，我们表明，并非所有知识都会很好地转移或对相关任务产生积极影响，即使它们共享一个共同的目标也是如此。我们基于数百种分为4组的视觉和语言任务进行了数百个跨表现的分析。尽管同一组中的任务容易相互改进，但结果表明并非总是如此。其他因素（例如数据集大小或训练阶段）也对知识的转移程度也有很大的影响。

Much of vision-and-language research focuses on a small but diverse set of independent tasks and supporting datasets often studied in isolation; however, the visuallygrounded language understanding skills required for success at these tasks overlap significantly. In this work, we investigate these relationships between vision-and-language tasks by developing a large-scale, multi-task training regime. Our approach culminates in a single model on 12 datasets from four broad categories of task including visual question answering, caption-based image retrieval, grounding referring expressions, and multi-modal verification. Compared to independently trained single-task models, this represents a reduction from approximately 3 billion parameters to 270 million while simultaneously improving performance by 2.05 points on average across tasks. We use our multi-task framework to perform in-depth analysis of the effect of joint training diverse tasks. Further, we show that finetuning task-specific models from our single multi-task model can lead to further improvements, achieving performance at or above the state-of-the-art.

We present Answer-Me, a task-aware multi-task framework which unifies a variety of question answering tasks, such as, visual question answering, visual entailment, visual reasoning. In contrast to previous works using contrastive or generative captioning training, we propose a novel and simple recipe to pre-train a vision-language joint model, which is multi-task as well. The pre-training uses only noisy image captioning data, and is formulated to use the entire architecture end-to-end with both a strong language encoder and decoder. Our results show state-of-the-art performance, zero-shot generalization, robustness to forgetting, and competitive single-task results across a variety of question answering tasks. Our multi-task mixture training learns from tasks of various question intents and thus generalizes better, including on zero-shot vision-language tasks. We conduct experiments in the challenging multi-task and open-vocabulary settings and across a variety of datasets and tasks, such as VQA2.0, SNLI-VE, NLVR2, GQA. We observe that the proposed approach is able to generalize to unseen tasks and that more diverse mixtures lead to higher accuracy in both known and novel tasks.

This paper presents a detailed study of improving visual representations for vision language (VL) tasks and develops an improved object detection model to provide object-centric representations of images. Compared to the most widely used bottom-up and top-down model [2], the new model is bigger, better-designed for VL tasks, and pre-trained on much larger training corpora that combine multiple public annotated object detection datasets. Therefore, it can generate representations of a richer collection of visual objects and concepts. While previous VL research focuses mainly on improving the vision-language fusion model and leaves the object detection model improvement untouched, we show that visual features matter significantly in VL models. In our experiments we feed the visual features generated by the new object detection model into a Transformer-based VL fusion model OSCAR [21], and utilize an improved approach OSCAR+ to pre-train the VL model and fine-tune it on a wide range of downstream VL tasks. Our results show that the new visual features significantly improve the performance across all VL tasks, creating new state-of-the-art results on seven public benchmarks. Code, models and pre-extracted features are released at https://github.com/pzzhang/VinVL. ♥ Microsoft Corporation♠ University of Washington † indicates equal contributions.

The availability of large-scale image captioning and visual question answering datasets has contributed significantly to recent successes in vision-and-language pretraining. However, these datasets are often collected with overrestrictive requirements inherited from their original target tasks (e.g., image caption generation), which limit the resulting dataset scale and diversity. We take a step further in pushing the limits of vision-and-language pretraining data by relaxing the data collection pipeline used in Conceptual Captions 3M (CC3M) [70] and introduce the Conceptual 12M (CC12M), a dataset with 12 million image-text pairs specifically meant to be used for visionand-language pre-training. We perform an analysis of this dataset and benchmark its effectiveness against CC3M on multiple downstream tasks with an emphasis on long-tail visual recognition. Our results clearly illustrate the benefit of scaling up pre-training data for vision-and-language tasks, as indicated by the new state-of-the-art results on both the nocaps and Conceptual Captions benchmarks. 1

We present ViLBERT (short for Vision-and-Language BERT), a model for learning task-agnostic joint representations of image content and natural language. We extend the popular BERT architecture to a multi-modal two-stream model, processing both visual and textual inputs in separate streams that interact through co-attentional transformer layers. We pretrain our model through two proxy tasks on the large, automatically collected Conceptual Captions dataset and then transfer it to multiple established vision-and-language tasks -visual question answering, visual commonsense reasoning, referring expressions, and caption-based image retrieval -by making only minor additions to the base architecture. We observe significant improvements across tasks compared to existing task-specific modelsachieving state-of-the-art on all four tasks. Our work represents a shift away from learning groundings between vision and language only as part of task training and towards treating visual grounding as a pretrainable and transferable capability.Preprint. Under review.

本文介绍了Omnivl，这是一种新的基础模型，旨在使用一种通用体系结构来支持图像语言和视频语言任务。它为图像和视频输入采用了统一的基于变压器的视觉编码器，因此可以执行联合图像语言和视频语言预处理。我们首次证明了这样的范式受益于图像和视频任务，而不是传统的单向传输（例如，使用图像语言来帮助视频语言）。为此，我们提出了对图像语言和视频语言的脱钩关节预处理，以有效地将视觉模型分解为空间和时间维度，并在图像和视频任务上获得性能提升。此外，我们引入了一种新颖的统一视觉对比度（UNIVLC）损失，以利用图像文本，视频文本，图像标签（例如，图像分类），视频标签（例如，视频动作识别）在一起受到监督和吵闹的监督预处理数据都尽可能多地利用。无需额外的任务适配器，Omnivl可以同时支持仅视觉任务（例如，图像分类，视频操作识别），跨模式对齐任务（例如，图像/视频 - 文本检索）和多模式理解和生成任务（例如，图像/视频问答，字幕）。我们在各种下游任务上评估Omnivl，并以相似的模型大小和数据量表获得最新的或竞争结果。

Joint image-text embedding is the bedrock for most Visionand-Language (V+L) tasks, where multimodality inputs are simultaneously processed for joint visual and textual understanding. In this paper, we introduce UNITER, a UNiversal Image-TExt Representation, learned through large-scale pre-training over four image-text datasets (COCO, Visual Genome, Conceptual Captions, and SBU Captions), which can power heterogeneous downstream V+L tasks with joint multimodal embeddings. We design four pre-training tasks: Masked Language Modeling (MLM), Masked Region Modeling (MRM, with three variants), Image-Text Matching (ITM), and Word-Region Alignment (WRA). Different from previous work that applies joint random masking to both modalities, we use conditional masking on pre-training tasks (i.e., masked language/region modeling is conditioned on full observation of image/text). In addition to ITM for global image-text alignment, we also propose WRA via the use of Optimal Transport (OT) to explicitly encourage finegrained alignment between words and image regions during pre-training. Comprehensive analysis shows that both conditional masking and OTbased WRA contribute to better pre-training. We also conduct a thorough ablation study to find an optimal combination of pre-training tasks. Extensive experiments show that UNITER achieves new state of the art across six V+L tasks (over nine datasets), including Visual Question

This paper presents a unified Vision-Language Pre-training (VLP) model. The model is unified in that (1) it can be finetuned for either vision-language generation (e.g., image captioning) or understanding (e.g., visual question answering) tasks, and (2) it uses a shared multi-layer transformer network for both encoding and decoding, which differs from many existing methods where the encoder and decoder are implemented using separate models. The unified VLP model is pre-trained on a large amount of image-text pairs using the unsupervised learning objectives of two tasks: bidirectional and sequence-to-sequence (seq2seq) masked vision-language prediction. The two tasks differ solely in what context the prediction conditions on. This is controlled by utilizing specific self-attention masks for the shared transformer network. To the best of our knowledge, VLP is the first reported model that achieves state-of-the-art results on both vision-language generation and understanding tasks, as disparate as image captioning and visual question answering, across three challenging benchmark datasets: COCO Captions, Flickr30k Captions, and VQA 2.0. The code and the pre-trained models are available at https://github.com/LuoweiZhou/VLP.

通用视觉（GPV）系统是旨在解决各种视觉任务的模型，而无需进行架构更改。如今，GPV主要从大型完全监督的数据集中学习技能和概念。通过获取数据以迅速学习每个技能的每个概念，将GPV扩展到数万个概念都变得令人望而却步。这项工作提出了一种有效且廉价的替代方法：从监督数据集中学习技能，从Web图像搜索中学习概念，并利用GPV的关键特征：跨技能传递视觉知识的能力。我们使用跨越10K+视觉概念的1M+图像的数据集来演示3个基准上的两个现有GPV（GPV-1和VL-T5）的Webly Supumented概念扩展：5个基于可可的数据集（80个主要概念），这是一个新的策划系列，这是一个新的策划系列。基于OpenImages和VisualGenome存储库（〜500个概念）以及Web衍生的数据集（10K+概念）的5个数据集。我们还提出了一种新的体系结构GPV-2，该架构支持各种任务 - 从分类和本地化等视觉任务到Qu Viewer+语言任务，例如QA和字幕，再到更多的利基市场，例如人类对象互动检测。 GPV-2从Web数据中受益匪浅，并且在这些基准测试中胜过GPV-1和VL-T5。我们的数据，代码和Web演示可在https://prior.allenai.org/projects/gpv2上获得。

近年来，统一的视觉语言框架已经大大提高，其中大多数采用编码器架构将图像文本任务统一为序列到序列的生成。但是，现有的视频语言（VIDL）模型仍需要在每个任务的模型体系结构和培训目标中进行特定于任务的设计。在这项工作中，我们探索了一个统一的VIDL框架薰衣草，其中蒙版语言建模（MLM）用作所有前训练和下游任务的常见接口。这样的统一导致了简化的模型体系结构，在多模式编码器之上，只需要一个轻巧的MLM头，而不是具有更多参数的解码器。令人惊讶的是，实验结果表明，这个统一的框架在14个VIDL基准测试中实现了竞争性能，涵盖了视频问答，文本到视频检索和视频字幕。广泛的分析进一步证明了薰衣草比现有VIDL方法的优势：（i）在多任务列出时仅使用一组参数值支持所有下游任务； （ii）对各种下游任务的几乎没有概括； （iii）在视频问题回答任务上启用零射门评估。代码可从https://github.com/microsoft/lavender获得。

Large-scale pre-training methods of learning cross-modal representations on image-text pairs are becoming popular for vision-language tasks. While existing methods simply concatenate image region features and text features as input to the model to be pre-trained and use selfattention to learn image-text semantic alignments in a brute force manner, in this paper, we propose a new learning method Oscar 1 , which uses object tags detected in images as anchor points to significantly ease the learning of alignments. Our method is motivated by the observation that the salient objects in an image can be accurately detected, and are often mentioned in the paired text. We pre-train an Oscar model on the public corpus of 6.5 million text-image pairs, and fine-tune it on downstream tasks, creating new state-of-the-arts on six well-established vision-language understanding and generation tasks. 2

我们提出了GLIPV2，这是一个接地的VL理解模型，该模型既服务于本地化任务（例如，对象检测，实例分割）和视觉语言（VL）理解任务（例如VQA，图像字幕）。 GLIPV2优雅地将本地化预训练和视觉语言预训练（VLP）具有三个预训练任务：短语接地作为对检测任务的VL重新重新制定，区域词对比度学习作为新型的区域词对比度对比度对比学习任务，以及蒙面的语言建模。这种统一不仅简化了先前的多阶段VLP程序，而且还可以在本地化和理解任务之间实现相互利益。实验结果表明，在各种本地化和理解任务上，单个GLIPV2模型（所有模型权重）在SOTA性能附近实现。该模型还显示了（1）在开放式摄制对象检测任务上进行的强零射击和很少的自适应性能，以及（2）VL理解任务上的卓越接地能力。代码将在https://github.com/microsoft/glip上发布。

视觉问题应答（VQA）任务利用视觉图像和语言分析来回回答图像的文本问题。它是一个流行的研究课题，在过去十年中越来越多的现实应用。本文介绍了我们最近对AliceMind-MMU的研究（阿里巴巴的编码器 - 解码器来自Damo Academy - 多媒体理解的机器智能实验室），其比人类在VQA上获得相似甚至略微更好的结果。这是通过系统地改善VQA流水线来实现的，包括：（1）具有全面的视觉和文本特征表示的预培训; （2）与学习参加的有效跨模型互动; （3）一个新颖的知识挖掘框架，具有专门的专业专家模块，适用于复杂的VQA任务。处理不同类型的视觉问题，需要具有相应的专业知识在提高我们的VQA架构的表现方面发挥着重要作用，这取决于人力水平。进行了广泛的实验和分析，以证明新的研究工作的有效性。

视觉问题回答是自然语言和愿景理解的重要任务。但是，在大多数公众视觉问题上回答了诸如VQA，CLEVR之类的数据集，这些问题是针对给定图像的特定于“她的眼睛是什么颜色？”的人类产生的。人类产生的众包问题相对简单，有时对某些实体或属性有偏见。在本文中，我们介绍了一个基于Image-Chiqa的新问题回答数据集。它包含Internet用户发布的现实查询，并结合了几个相关的开放域图像。系统应确定图像是否可以回答问题。与以前的VQA数据集不同，这些问题是现实世界中独立的查询，这些查询更加各种和无偏见。与先前的图像回程或图像捕获数据集相比，Chiqa不仅衡量了相关性，而且还可以衡量答案性，这需要更细粒度的视力和语言推理。 Chiqa包含超过40k的问题和超过200k的问题图像对。将三级2/1/0标签分配给每个对，指示完美的答案，部分答案和无关紧要。数据分析表明，Chiqa需要对语言和视觉有深入的了解，包括接地，比较和阅读。我们评估了几种最先进的视觉语言模型，例如ALBEF，表明仍然有一个很大的改进奇卡的空间。

We study joint learning of Convolutional Neural Network (CNN) and Transformer for vision-language pre-training (VLPT) which aims to learn cross-modal alignments from millions of image-text pairs. State-of-the-art approaches extract salient image regions and align regions with words step-by-step. As region-based visual features usually represent parts of an image, it is challenging for existing visionlanguage models to fully understand the semantics from paired natural languages. In this paper, we propose SOHO to "See Out of tHe bOx" that takes a whole image as input, and learns vision-language representation in an endto-end manner. SOHO does not require bounding box annotations which enables inference 10 times faster than regionbased approaches. In particular, SOHO learns to extract comprehensive yet compact image features through a visual dictionary (VD) that facilitates cross-modal understanding. VD is designed to represent consistent visual abstractions of similar semantics. It is updated on-the-fly and utilized in our proposed pre-training task Masked Visual Modeling (MVM). We conduct experiments on four well-established vision-language tasks by following standard VLPT settings. In particular, SOHO achieves absolute gains of 2.0% R@1 score on MSCOCO text retrieval 5k test split, 1.5% accuracy on NLVR 2 test-P split, 6.7% accuracy on SNLI-VE test split, respectively.

Large-scale vision-language models such as CLIP have shown impressive performance on zero-shot image classification and image-to-text retrieval. However, such zero-shot performance of CLIP-based models does not realize in tasks that require a finer-grained correspondence between vision and language, such as Visual Question Answering (VQA). We investigate why this is the case, and report an interesting phenomenon of CLIP, which we call the Concept Association Bias (CAB), as a potential cause of the difficulty of applying CLIP to VQA and similar tasks. CAB is especially apparent when two concepts are present in the given image while a text prompt only contains a single concept. In such a case, we find that CLIP tends to treat input as a bag of concepts and attempts to fill in the other missing concept crossmodally, leading to an unexpected zero-shot prediction. For example, when asked for the color of a lemon in an image, CLIP predicts ``purple'' if the image contains a lemon and an eggplant. We demonstrate the Concept Association Bias of CLIP by showing that CLIP's zero-shot classification performance greatly suffers when there is a strong concept association between an object (e.g. lemon) and an attribute (e.g. its color). On the other hand, when the association between object and attribute is weak, we do not see this phenomenon. Furthermore, we show that CAB is significantly mitigated when we enable CLIP to learn deeper structure across image and text embeddings by adding an additional Transformer on top of CLIP and fine-tuning it on VQA. We find that across such fine-tuned variants of CLIP, the strength of CAB in a model predicts how well it performs on VQA.

有效的缩放和灵活的任务接口使大型语言模型能够在许多任务中表现出色。帕利（Pali）根据视觉和文本输入生成文本，并使用该界面以许多语言执行许多视觉，语言和多模式任务。为了训练帕利，我们利用了大型的编码器语言模型和视觉变压器（VITS）。这使我们能够利用其现有能力，并利用培训它们的大量成本。我们发现，视觉和语言组成部分的联合缩放很重要。由于现有的语言变压器比其视觉对应物要大得多，因此我们训练迄今为止最大的VIT（VIT-E），以量化甚至大容量视觉模型的好处。为了训练Pali，我们基于一个新的图像文本训练集，其中包含10B图像和文本，以100多种语言来创建大型的多语言组合。帕利（Pali）在多个视觉和语言任务（例如字幕，视觉问题，索方式，场景文本理解）中实现了最新的，同时保留了简单，模块化和可扩展的设计。

Vision-and-language reasoning requires an understanding of visual concepts, language semantics, and, most importantly, the alignment and relationships between these two modalities. We thus propose the LXMERT (Learning Cross-Modality Encoder Representations from Transformers) framework to learn these vision-and-language connections. In LXMERT, we build a large-scale Transformer model that consists of three encoders: an object relationship encoder, a language encoder, and a cross-modality encoder. Next, to endow our model with the capability of connecting vision and language semantics, we pre-train the model with large amounts of image-and-sentence pairs, via five diverse representative pre-training tasks: masked language modeling, masked object prediction (feature regression and label classification), cross-modality matching, and image question answering. These tasks help in learning both intra-modality and cross-modality relationships. After fine-tuning from our pretrained parameters, our model achieves the state-of-the-art results on two visual question answering datasets (i.e., VQA and GQA). We also show the generalizability of our pretrained cross-modality model by adapting it to a challenging visual-reasoning task, NLVR 2 , and improve the previous best result by 22% absolute (54% to 76%). Lastly, we demonstrate detailed ablation studies to prove that both our novel model components and pretraining strategies significantly contribute to our strong results; and also present several attention visualizations for the different encoders. 1

This paper focuses on analyzing and improving the commonsense ability of recent popular vision-language (VL) models. Despite the great success, we observe that existing VL-models still lack commonsense knowledge/reasoning ability (e.g., "Lemons are sour"), which is a vital component towards artificial general intelligence. Through our analysis, we find one important reason is that existing large-scale VL datasets do not contain much commonsense knowledge, which motivates us to improve the commonsense of VL-models from the data perspective. Rather than collecting a new VL training dataset, we propose a more scalable strategy, i.e., "Data Augmentation with kNowledge graph linearization for CommonsensE capability" (DANCE). It can be viewed as one type of data augmentation technique, which can inject commonsense knowledge into existing VL datasets on the fly during training. More specifically, we leverage the commonsense knowledge graph (e.g., ConceptNet) and create variants of text description in VL datasets via bidirectional sub-graph sequentialization. For better commonsense evaluation, we further propose the first retrieval-based commonsense diagnostic benchmark. By conducting extensive experiments on some representative VL-models, we demonstrate that our DANCE technique is able to significantly improve the commonsense ability while maintaining the performance on vanilla retrieval tasks. The code and data are available at https://github.com/pleaseconnectwifi/DANCE