This paper introduces the ColorSwap dataset, designed to assess and improve the proficiency of multimodal models in matching objects with their colors. The dataset is comprised of 2,000 unique image-caption pairs, grouped into 1,000 examples. Each example includes a caption-image pair, along with a ``color-swapped'' pair. We follow the Winoground schema: the two captions in an example have the same words, but the color words have been rearranged to modify different objects. The dataset was created through a novel blend of automated caption and image generation with humans in the loop. We evaluate image-text matching (ITM) and visual language models (VLMs) and find that even the latest ones are still not robust at this task. GPT-4V and LLaVA score 72% and 42% on our main VLM metric, although they may improve with more advanced prompting techniques. On the main ITM metric, contrastive models such as CLIP and SigLIP perform close to chance (at 12% and 30%, respectively), although the non-contrastive BLIP ITM model is stronger (87%). We also find that finetuning on fewer than 2,000 examples yields significant performance gains on this out-of-distribution word-order understanding task. The dataset is here: https://github.com/Top34051/colorswap and here: https://huggingface.co/datasets/stanfordnlp/colorswap.
Beihang University（北京航空航天大学）；Alibaba Group（阿里巴巴集团）；University of Adelaide（阿德莱德大学）
When describing an image, reading text in the visual scene is crucial to understand the key information. Recent work explores the TextCaps task, i.e. image captioning with reading Optical Character Recognition (OCR) tokens, which requires models to read text and cover them in generated captions. Existing approaches fail to generate accurate descriptions because of their (1) poor reading ability; (2) inability to choose the crucial words among all extracted OCR tokens; (3) repetition of words in predicted captions. To this end, we propose a Confidence-aware Non-repetitive Multimodal Transformer…
Transformer models trained on massive text corpora have become the de facto models for a wide range of natural language processing tasks. However, learning effective word representations for function words remains challenging. Multimodal learning, which visually grounds transformer models in imagery, can overcome the challenges to some extent; however, there is still much work to be done. In this study, we explore the extent to which visual grounding facilitates the acquisition of function words using stable diffusion models that employ multimodal models for text-to-image generation. Out of se…
Is a Picture Worth a Thousand Words? Adaptive Multimodal Fact-Checking with Visual Evidence Necessity
School of AI Convergence, Soongsil University（顺斯利大学人工智能融合学院）；MAUM AI Inc.（MAUM人工智能公司）；Department of Intelligent Semiconductors, Soongsil University（顺斯利大学智能半导体系）
Assessing the alignment between infants' visual and linguistic experience using multimodal language models
Department of Psychology, Stanford University（心理学系，斯坦福大学）；Department of Psychology, University of California, San Diego（心理学系，加州大学圣地亚哥分校）；Department of Psychology, Tsinghua University（心理学系，清华大学）
Sun Yat-sen University（中山大学）；Microsoft Research Asia（微软亚洲研究院）
Self-supervised pre-training techniques have achieved remarkable progress in Document AI. Most multimodal pre-trained models use a masked language modeling objective to learn bidirectional representations on the text modality, but they differ in pre-training objectives for the image modality. This discrepancy adds difficulty to multimodal representation learning. In this paper, we propose \textbf{LayoutLMv3} to pre-train multimodal Transformers for Document AI with unified text and image masking. Additionally, LayoutLMv3 is pre-trained with a word-patch alignment objective to learn cross-modal…
Text-to-image multimodal tasks, generating/retrieving an image from a given text description, are extremely challenging tasks since raw text descriptions cover quite limited information in order to fully describe visually realistic images. We propose a new visual contextual text representation for text-to-image multimodal tasks, VICTR, which captures rich visual semantic information of objects from the text input. First, we use the text description as initial input and conduct dependency parsing to extract the syntactic structure and analyse the semantic aspect, including object quantities, to…
When trained at sufficient scale, auto-regressive language models exhibit the notable ability to learn a new language task after being prompted with just a few examples. Here, we present a simple, yet effective, approach for transferring this few-shot learning ability to a multimodal setting (vision and language). Using aligned image and caption data, we train a vision encoder to represent each image as a sequence of continuous embeddings, such that a pre-trained, frozen language model prompted with this prefix generates the appropriate caption. The resulting system is a multimodal few-shot le…
Renmin University of China（中国人民大学）；Carnegie Mellon University（卡内基梅隆大学）
Bilingual lexicon induction, translating words from the source language to the target language, is a long-standing natural language processing task. Recent endeavors prove that it is promising to employ images as pivot to learn the lexicon induction without reliance on parallel corpora. However, these vision-based approaches simply associate words with entire images, which are constrained to translate concrete words and require object-centered images. We humans can understand words better when they are within a sentence with context. Therefore, in this paper, we propose to utilize images and t…
Due to the complex layouts of documents, it is challenging to extract information for documents. Most previous studies develop multimodal pre-trained models in a self-supervised way. In this paper, we focus on the embedding learning of word blocks containing text and layout information, and propose UTel, a language model with Unified TExt and Layout pre-training. Specifically, we propose two pre-training tasks: Surrounding Word Prediction (SWP) for the layout learning, and Contrastive learning of Word Embeddings (CWE) for identifying different word blocks. Moreover, we replace the commonly use…
University of California, Los Angeles（加利福尼亚大学洛杉矶分校）；Pinterest Inc.（Pinterest公司）；Johns Hopkins University（约翰斯·霍普金斯大学）
In this paper, we focus on training and evaluating effective word embeddings with both text and visual information. More specifically, we introduce a large-scale dataset with 300 million sentences describing over 40 million images crawled and downloaded from publicly available Pins (i.e. an image with sentence descriptions uploaded by users) on Pinterest. This dataset is more than 200 times larger than MS COCO, the standard large-scale image dataset with sentence descriptions. In addition, we construct an evaluation dataset to directly assess the effectiveness of word embeddings in terms of fi…
Multimodal machine learning algorithms aim to learn visual-textual correspondences. Previous work suggests that concepts with concrete visual manifestations may be easier to learn than concepts with abstract ones. We give an algorithm for automatically computing the visual concreteness of words and topics within multimodal datasets. We apply the approach in four settings, ranging from image captions to images/text scraped from historical books. In addition to enabling explorations of concepts in multimodal datasets, our concreteness scores predict the capacity of machine learning algorithms to…
MEWL: Few-shot multimodal word learning with referential uncertainty
Peking University（北京大学）；Beijing Institute for General Artificial Intelligence（北京通用人工智能研究院）；Beijing Institute of Technology（北京理工大学）；Yangtze Delta Region Academy of Beijing Institute of Technology（北京理工大学长三角研究院）
Online conversations include more than just text. Increasingly, image-based responses such as memes and animated gifs serve as culturally recognized and often humorous responses in conversation. However, while NLP has broadened to multimodal models, conversational dialog systems have largely focused only on generating text replies. Here, we introduce a new dataset of 1.56M text-gif conversation turns and introduce a new multimodal conversational model Pepe the King Prawn for selecting gif-based replies. We demonstrate that our model produces relevant and high-quality gif responses and, in a la…
This work studies the representational mapping across multimodal data such that given a piece of the raw data in one modality the corresponding semantic description in terms of the raw data in another modality is immediately obtained. Such a representational mapping can be found in a wide spectrum of real-world applications including image/video retrieval, object recognition, action/behavior recognition, and event understanding and prediction. To that end, we introduce a simplified training objective for learning multimodal embeddings using the skip-gram architecture by introducing convolution…
Shanghai Jiao Tong University（上海交通大学）；National Institute of Information（国立信息研究所）
Word representation is a fundamental component in neural language understanding models. Recently, pre-trained language models (PrLMs) offer a new performant method of contextualized word representations by leveraging the sequence-level context for modeling. Although the PrLMs generally give more accurate contextualized word representations than non-contextualized models do, they are still subject to a sequence of text contexts without diverse hints for word representation from multimodality. This paper thus proposes a visual representation method to explicitly enhance conventional word embeddi…
Center for Mind/Brain Sciences（心智/脑科学中心）；University of Trento（特伦托大学）
We extend the SKIP-GRAM model of Mikolov et al. (2013a) by taking visual information into account. Like SKIP-GRAM, our multimodal models (MMSKIP-GRAM) build vector-based word representations by learning to predict linguistic contexts in text corpora. However, for a restricted set of words, the models are also exposed to visual representations of the objects they denote (extracted from natural images), and must predict linguistic and visual features jointly. The MMSKIP-GRAM models achieve good performance on a variety of semantic benchmarks. Moreover, since they propagate visual information to …
KAIST（韩国科学技术院）；Technical University of Munich（慕尼黑工业大学）
The ambiguity of the decision-making process has been pointed out as the main obstacle to applying the deep learning-based method in a practical way in spite of its outstanding performance. Interpretability could guarantee the confidence of deep learning system, therefore it is particularly important in the medical field. In this study, a novel deep network is proposed to explain the diagnostic decision with visual pointing map and diagnostic sentence justifying result simultaneously. For the purpose of increasing the accuracy of sentence generation, a visual word constraint model is devised i…
In this paper we are interested in the problem of image segmentation given natural language descriptions, i.e. referring expressions. Existing works tackle this problem by first modeling images and sentences independently and then segment images by combining these two types of representations. We argue that learning word-to-image interaction is more native in the sense of jointly modeling two modalities for the image segmentation task, and we propose convolutional multimodal LSTM to encode the sequential interactions between individual words, visual information, and spatial information. We sho…
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）；Institute of Artificial Intelligence, Beihang University（北京航空航天大学人工智能研究院）；Sun Yat-sen University（中山大学）；Tencent Marketing Solution（腾讯营销解决方案）
Referring image segmentation aims to predict the foreground mask of the object referred by a natural language sentence. Multimodal context of the sentence is crucial to distinguish the referent from the background. Existing methods either insufficiently or redundantly model the multimodal context. To tackle this problem, we propose a "gather-propagate-distribute" scheme to model multimodal context by cross-modal interaction and implement this scheme as a novel Linguistic Structure guided Context Modeling (LSCM) module. Our LSCM module builds a Dependency Parsing Tree suppressed Word Graph (DPT…
Bridging Pixels and Words: Mask-Aware Local Semantic Fusion for Multimodal Media Verification
State Key Laboratory of Human-Machine Hybrid Augmented Intelligence, Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University（西安交通大学人工智能与机器人研究所人机混合增强智能全国重点实验室）
Baidu Research（百度研究院）；University of California, Los Angeles（加州大学洛杉矶分校）
In this paper, we present a multimodal Recurrent Neural Network (m-RNN) model for generating novel sentence descriptions to explain the content of images. It directly models the probability distribution of generating a word given previous words and the image. Image descriptions are generated by sampling from this distribution. The model consists of two sub-networks: a deep recurrent neural network for sentences and a deep convolutional network for images. These two sub-networks interact with each other in a multimodal layer to form the whole m-RNN model. The effectiveness of our model is valid…
In this paper, we propose multimodal convolutional neural networks (m-CNNs) for matching image and sentence. Our m-CNN provides an end-to-end framework with convolutional architectures to exploit image representation, word composition, and the matching relations between the two modalities. More specifically, it consists of one image CNN encoding the image content, and one matching CNN learning the joint representation of image and sentence. The matching CNN composes words to different semantic fragments and learns the inter-modal relations between image and the composed fragments at different …
College of Intelligence and Computing, Tianjin University（天津大学智能与计算学部）；School of New Media and Communication, Tianjin University（天津大学新媒体与传播学院）；College of Artificial Intelligence, Tianjin University of Science and Technology（天津科技大学人工智能学院）
With massive explosion of social media such as Twitter and Instagram, people daily share billions of multimedia posts, containing images and text. Typically, text in these posts is short, informal and noisy, leading to ambiguities which can be resolved using images. In this paper we explore text-centric Named Entity Recognition task on these multimedia posts. We propose an end to end model which learns a joint representation of a text and an image. Our model extends multi-dimensional self attention technique, where now image helps to enhance relationship between words. Experiments show that ou…
Aix Marseille Univ（艾克斯-马赛大学）；Université de Toulon（土伦大学）；CNRS（法国国家科学研究中心）
Recent advances in vision-and-language modeling have seen the development of Transformer architectures that achieve remarkable performance on multimodal reasoning tasks. Yet, the exact capabilities of these black-box models are still poorly understood. While much of previous work has focused on studying their ability to learn meaning at the word-level, their ability to track syntactic dependencies between words has received less attention. We take a first step in closing this gap by creating a new multimodal task targeted at evaluating understanding of predicate-noun dependencies in a controll…
GEM-VPC: A dual Graph-Enhanced Multimodal integration for Video Paragraph Captioning
School of Computer Science, The University of Sydney（悉尼大学计算机科学学院）；School of Computing and Information Systems, The University of Melbourne（墨尔本大学计算与信息系统学院）
Institute for Language, Cognition and Computation（语言、认知与计算研究所）；School of Informatics, University of Edinburgh（爱丁堡大学信息学院）
We introduce a new task, visual sense disambiguation for verbs: given an image and a verb, assign the correct sense of the verb, i.e., the one that describes the action depicted in the image. Just as textual word sense disambiguation is useful for a wide range of NLP tasks, visual sense disambiguation can be useful for multimodal tasks such as image retrieval, image description, and text illustration. We introduce VerSe, a new dataset that augments existing multimodal datasets (COCO and TUHOI) with sense labels. We propose an unsupervised algorithm based on Lesk which performs visual sense dis…
University of California, Los Angeles（加州大学洛杉矶分校）；Baidu Research（百度研究院）
In this paper, we present a multimodal Recurrent Neural Network (m-RNN) model for generating novel image captions. It directly models the probability distribution of generating a word given previous words and an image. Image captions are generated by sampling from this distribution. The model consists of two sub-networks: a deep recurrent neural network for sentences and a deep convolutional network for images. These two sub-networks interact with each other in a multimodal layer to form the whole m-RNN model. The effectiveness of our model is validated on four benchmark datasets (IAPR TC-12, …
University of Michigan（密歇根大学）；National University of Singapore（新加坡国立大学）；SUTD（新加坡科技设计大学）
Sarcasm is often expressed through several verbal and non-verbal cues, e.g., a change of tone, overemphasis in a word, a drawn-out syllable, or a straight looking face. Most of the recent work in sarcasm detection has been carried out on textual data. In this paper, we argue that incorporating multimodal cues can improve the automatic classification of sarcasm. As a first step towards enabling the development of multimodal approaches for sarcasm detection, we propose a new sarcasm dataset, Multimodal Sarcasm Detection Dataset (MUStARD), compiled from popular TV shows. MUStARD consists of audio…
This paper is a technical report to share our experience and findings building a Korean and English bilingual multimodal model. While many of the multimodal datasets focus on English and multilingual multimodal research uses machine-translated texts, employing such machine-translated texts is limited to describing unique expressions, cultural information, and proper noun in languages other than English. In this work, we collect 1.1 billion image-text pairs (708 million Korean and 476 million English) and train a bilingual multimodal model named KELIP. We introduce simple yet effective training…
Large datasets as required for deep learning of lip reading do not exist in many languages. In this paper we present the dataset GLips (German Lips) consisting of 250,000 publicly available videos of the faces of speakers of the Hessian Parliament, which was processed for word-level lip reading using an automatic pipeline. The format is similar to that of the English language LRW (Lip Reading in the Wild) dataset, with each video encoding one word of interest in a context of 1.16 seconds duration, which yields compatibility for studying transfer learning between both datasets. By training a de…
Microsoft Research（微软研究院）；University of Washington（华盛顿大学）；University of California at Berkeley（加州大学伯克利分校）；IDSIA（IDSIA研究所）；USI-SUPSI（意大利语大学-瑞士南部应用科学与艺术大学）；Facebook AI Research（脸书人工智能研究院）；Google（谷歌）
This paper presents a novel approach for automatically generating image descriptions: visual detectors, language models, and multimodal similarity models learnt directly from a dataset of image captions. We use multiple instance learning to train visual detectors for words that commonly occur in captions, including many different parts of speech such as nouns, verbs, and adjectives. The word detector outputs serve as conditional inputs to a maximum-entropy language model. The language model learns from a set of over 400,000 image descriptions to capture the statistics of word usage. We capture…
MIT Computer Science and Artificial Intelligence Laboratory（麻省理工学院计算机科学与人工智能实验室）
In this paper, we present a model which takes as input a corpus of images with relevant spoken captions and finds a correspondence between the two modalities. We employ a pair of convolutional neural networks to model visual objects and speech signals at the word level, and tie the networks together with an embedding and alignment model which learns a joint semantic space over both modalities. We evaluate our model using image search and annotation tasks on the Flickr8k dataset, which we augmented by collecting a corpus of 40,000 spoken captions using Amazon Mechanical Turk.
We address the problem of phrase grounding by lear ing a multi-level common semantic space shared by the textual and visual modalities. We exploit multiple levels of feature maps of a Deep Convolutional Neural Network, as well as contextualized word and sentence embeddings extracted from a character-based language model. Following dedicated non-linear mappings for visual features at each level, word, and sentence embeddings, we obtain multiple instantiations of our common semantic space in which comparisons between any target text and the visual content is performed with cosine similarity. We …
We propose MoodNet - A Deep Convolutional Neural Network based architecture to effectively predict the emotion associated with a piece of music given its audio and lyrical content.We evaluate different architectures consisting of varying number of two-dimensional convolutional and subsampling layers,followed by dense layers.We use Mel-Spectrograms to represent the audio content and word embeddings-specifically 100 dimensional word vectors, to represent the textual content represented by the lyrics.We feed input data from both modalities to our MoodNet architecture.The output from both the moda…
We propose a new spatial memory module and a spatial reasoner for the Visual Grounding (VG) task. The goal of this task is to find a certain object in an image based on a given textual query. Our work focuses on integrating the regions of a Region Proposal Network (RPN) into a new multi-step reasoning model which we have named a Multimodal Spatial Region Reasoner (MSRR). The introduced model uses the object regions from an RPN as initialization of a 2D spatial memory and then implements a multi-step reasoning process scoring each region according to the query, hence why we call it a multimodal…
Traditionally, text generation models take in a sequence of text as input, and iteratively generate the next most probable word using pre-trained parameters. In this work, we propose the architecture to use images instead of text as the input of the text generation model, called StoryGen. In the architecture, we design a Relational Text Data Generator algorithm that relates different features from multiple images. The output samples from the model demonstrate the ability to generate meaningful paragraphs of text containing the extracted features from the input images. This is an undergraduate …
Tsinghua University（清华大学）；Toutiao AI Lab（今日头条人工智能实验室）；UESTC（电子科技大学）；Johns Hopkins University（约翰斯·霍普金斯大学）
In this work we formulate the problem of image captioning as a multimodal translation task. Analogous to machine translation, we present a sequence-to-sequence recurrent neural networks (RNN) model for image caption generation. Different from most existing work where the whole image is represented by convolutional neural network (CNN) feature, we propose to represent the input image as a sequence of detected objects which feeds as the source sequence of the RNN model. In this way, the sequential representation of an image can be naturally translated to a sequence of words, as the target sequen…
Classification of document images is a critical step for archival of old manuscripts, online subscription and administrative procedures. Computer vision and deep learning have been suggested as a first solution to classify documents based on their visual appearance. However, achieving the fine-grained classification that is required in real-world setting cannot be achieved by visual analysis alone. Often, the relevant information is in the actual text content of the document. We design a multimodal neural network that is able to learn from word embeddings, computed on text extracted by OCR, an…
We introduce a novel approach for scanned document representation to perform field extraction. It allows the simultaneous encoding of the textual, visual and layout information in a 3-axis tensor used as an input to a segmentation model. We improve the recent Chargrid and Wordgrid \cite{chargrid} models in several ways, first by taking into account the visual modality, then by boosting its robustness in regards to small datasets while keeping the inference time low. Our approach is tested on public and private document-image datasets, showing higher performances compared to the recent state-of…
Tsinghua University（清华大学）；Université de Sherbrooke（舍布鲁克大学）
Topic modeling based on latent Dirichlet allocation (LDA) has been a framework of choice to deal with multimodal data, such as in image annotation tasks. Another popular approach to model the multimodal data is through deep neural networks, such as the deep Boltzmann machine (DBM). Recently, a new type of topic model called the Document Neural Autoregressive Distribution Estimator (DocNADE) was proposed and demonstrated state-of-the-art performance for text document modeling. In this work, we show how to successfully apply and extend this model to multimodal data, such as simultaneous image cl…
