Visual Commonsense Reasoning (VCR) predicts an answer with corresponding rationale, given a question-image input. VCR is a recently introduced visual scene understanding task with a wide range of applications, including visual question answering, automated vehicle systems, and clinical decision support. Previous approaches to solving the VCR task generally rely on pre-training or exploiting memory with long dependency relationship encoded models. However, these approaches suffer from a lack of generalizability and prior knowledge. In this paper we propose a dynamic working memory based cognitive VCR network, which stores accumulated commonsense between sentences to provide prior knowledge for inference. Extensive experiments show that the proposed model yields significant improvements over existing methods on the benchmark VCR dataset. Moreover, the proposed model provides intuitive interpretation into visual commonsense reasoning. A Python implementation of our mechanism is publicly available at https://github.com/tanjatang/DMVCR
As humans, we understand events in the visual world contextually, performing multimodal reasoning across time to make inferences about the past, present, and future. We introduce MERLOT, a model that learns multimodal script knowledge by watching millions of YouTube videos with transcribed speech -- in an entirely label-free, self-supervised manner. By pretraining with a mix of both frame-level (spatial) and video-level (temporal) objectives, our model not only learns to match images to temporally corresponding words, but also to contextualize what is happening globally over time. As a result, MERLOT exhibits strong out-of-the-box representations of temporal commonsense, and achieves state-of-the-art performance on 12 different video QA datasets when finetuned. It also transfers well to the world of static images, allowing models to reason about the dynamic context behind visual scenes. On Visual Commonsense Reasoning, MERLOT answers questions correctly with 80.6% accuracy, outperforming state-of-the-art models of similar size by over 3%, even those that make heavy use of auxiliary supervised data (like object bounding boxes). Ablation analyses demonstrate the complementary importance of: 1) training on videos versus static images; 2) scaling the magnitude and diversity of the pretraining video corpus; and 3) using diverse objectives that encourage full-stack multimodal reasoning, from the recognition to cognition level.
It is a common practice for recent works in vision language cross-modal reasoning to adopt a binary or multi-choice classification formulation taking as input a set of source image(s) and textual query. In this work, we take a sober look at such an unconditional formulation in the sense that no prior knowledge is specified with respect to the source image(s). Inspired by the designs of both visual commonsense reasoning and natural language inference tasks, we propose a new task termed Premise-based Multi-modal Reasoning(PMR) where a textual premise is the background presumption on each source image. The PMR dataset contains 15,360 manually annotated samples which are created by a multi-phase crowd-sourcing process. With selected high-quality movie screenshots and human-curated premise templates from 6 pre-defined categories, we ask crowd-source workers to write one true hypothesis and three distractors (4 choices) given the premise and image through a cross-check procedure. Besides, we generate adversarial samples to alleviate the annotation artifacts and double the size of PMR. We benchmark various state-of-the-art (pretrained) multi-modal inference models on PMR and conduct comprehensive experimental analyses to showcase the utility of our dataset.
Large-scale pre-training has recently revolutionized vision-and-language (VL) research. Models such as LXMERT and UNITER have significantly lifted the state of the art over a wide range of VL tasks. However, the large number of parameters in such models hinders their application in practice. In parallel, work on the lottery ticket hypothesis (LTH) has shown that deep neural networks contain small matching subnetworks that can achieve on par or even better performance than the dense networks when trained in isolation. In this work, we perform the first empirical study to assess whether such trainable subnetworks also exist in pre-trained VL models. We use UNITER as the main testbed (also test on LXMERT and ViLT), and consolidate 7 representative VL tasks for experiments, including visual question answering, visual commonsense reasoning, visual entailment, referring expression comprehension, image-text retrieval, GQA, and NLVR$^2$. Through comprehensive analysis, we summarize our main findings as follows. ($i$) It is difficult to find subnetworks that strictly match the performance of the full model. However, we can find "relaxed" winning tickets at 50%-70% sparsity that maintain 99% of the full accuracy. ($ii$) Subnetworks found by task-specific pruning transfer reasonably well to the other tasks, while those found on the pre-training tasks at 60%/70% sparsity transfer universally, matching 98%/96% of the full accuracy on average over all the tasks. ($iii$) Besides UNITER, other models such as LXMERT and ViLT can also play lottery tickets. However, the highest sparsity we can achieve for ViLT is far lower than LXMERT and UNITER (30% vs. 70%). ($iv$) LTH also remains relevant when using other training methods (e.g., adversarial training).
Existing methods for vision-and-language learning typically require designing task-specific architectures and objectives for each task. For example, a multi-label answer classifier for visual question answering, a region scorer for referring expression comprehension, and a language decoder for image captioning, etc. To alleviate these hassles, in this work, we propose a unified framework that learns different tasks in a single architecture with the same language modeling objective, i.e., multimodal conditional text generation, where our models learn to generate labels in text based on the visual and textual inputs. On 7 popular vision-and-language benchmarks, including visual question answering, referring expression comprehension, visual commonsense reasoning, most of which have been previously modeled as discriminative tasks, our generative approach (with a single unified architecture) reaches comparable performance to recent task-specific state-of-the-art vision-and-language models. Moreover, our generative approach shows better generalization ability on questions that have rare answers. Also, we show that our framework allows multi-task learning in a single architecture with a single set of parameters, achieving similar performance to separately optimized single-task models. Our code is publicly available at: https://github.com/j-min/VL-T5
Reasoning is a critical ability towards complete visual understanding. To develop machine with cognition-level visual understanding and reasoning abilities, the visual commonsense reasoning (VCR) task has been introduced. In VCR, given a challenging question about an image, a machine must answer correctly and then provide a rationale justifying its answer. The methods adopting the powerful BERT model as the backbone for learning joint representation of image content and natural language have shown promising improvements on VCR. However, none of the existing methods have utilized commonsense knowledge in visual commonsense reasoning, which we believe will be greatly helpful in this task. With the support of commonsense knowledge, complex questions even if the required information is not depicted in the image can be answered with cognitive reasoning. Therefore, we incorporate commonsense knowledge into the cross-modal BERT, and propose a novel Knowledge Enhanced Visual-and-Linguistic BERT (KVL-BERT for short) model. Besides taking visual and linguistic contents as input, external commonsense knowledge extracted from ConceptNet is integrated into the multi-layer Transformer. In order to reserve the structural information and semantic representation of the original sentence, we propose using relative position embedding and mask-self-attention to weaken the effect between the injected commonsense knowledge and other unrelated components in the input sequence. Compared to other task-specific models and general task-agnostic pre-training models, our KVL-BERT outperforms them by a large margin.
Natural language rationales could provide intuitive, higher-level explanations that are easily understandable by humans, complementing the more broadly studied lower-level explanations based on gradients or attention weights. We present the first study focused on generating natural language rationales across several complex visual reasoning tasks: visual commonsense reasoning, visual-textual entailment, and visual question answering. The key challenge of accurate rationalization is comprehensive image understanding at all levels: not just their explicit content at the pixel level, but their contextual contents at the semantic and pragmatic levels. We present Rationale^VT Transformer, an integrated model that learns to generate free-text rationales by combining pretrained language models with object recognition, grounded visual semantic frames, and visual commonsense graphs. Our experiments show that the base pretrained language model benefits from visual adaptation and that free-text rationalization is a promising research direction to complement model interpretability for complex visual-textual reasoning tasks.
One of the ambitions of artificial intelligence is to root artificial intelligence deeply in basic science while developing brain-inspired artificial intelligence platforms that will promote new scientific discoveries. The challenges are essential to push artificial intelligence theory and applied technologies research forward. This paper presents the grand challenges of artificial intelligence research for the next 20 years which include:~(i) to explore the working mechanism of the human brain on the basis of understanding brain science, neuroscience, cognitive science, psychology and data science; (ii) how is the electrical signal transmitted by the human brain? What is the coordination mechanism between brain neural electrical signals and human activities? (iii)~to root brain-computer interface~(BCI) and brain-muscle interface~(BMI) technologies deeply in science on human behaviour; (iv)~making research on knowledge-driven visual commonsense reasoning~(VCR), develop a new inference engine for cognitive network recognition~(CNR); (v)~to develop high-precision, multi-modal intelligent perceptrons; (vi)~investigating intelligent reasoning and fast decision-making systems based on knowledge graph~(KG). We believe that the frontier theory innovation of AI, knowledge-driven modeling methodologies for commonsense reasoning, revolutionary innovation and breakthroughs of the novel algorithms and new technologies in AI, and developing responsible AI should be the main research strategies of AI scientists in the future.
We present VILLA, the first known effort on large-scale adversarial training for vision-and-language (V+L) representation learning. VILLA consists of two training stages: (i) task-agnostic adversarial pre-training; followed by (ii) task-specific adversarial finetuning. Instead of adding adversarial perturbations on image pixels and textual tokens, we propose to perform adversarial training in the embedding space of each modality. To enable large-scale training, we adopt the "free" adversarial training strategy, and combine it with KL-divergence-based regularization to promote higher invariance in the embedding space. We apply VILLA to current best-performing V+L models, and achieve new state of the art on a wide range of tasks, including Visual Question Answering, Visual Commonsense Reasoning, Image-Text Retrieval, Referring Expression Comprehension, Visual Entailment, and NLVR2.
Even from a single frame of a still image, people can reason about the dynamic story of the image before, after, and beyond the frame. For example, given an image of a man struggling to stay afloat in water, we can reason that the man fell into the water sometime in the past, the intent of that man at the moment is to stay alive, and he will need help in the near future or else he will get washed away. We propose VisualComet, the novel framework of visual commonsense reasoning tasks to predict events that might have happened before, events that might happen next, and the intents of the people at present. To support research toward visual commonsense reasoning, we introduce the first large-scale repository of Visual Commonsense Graphs that consists of over 1.4 million textual descriptions of visual commonsense inferences carefully annotated over a diverse set of 60,000 images, each paired with short video summaries of before and after. In addition, we provide person-grounding (i.e., co-reference links) between people appearing in the image and people mentioned in the textual commonsense descriptions, allowing for tighter integration between images and text. We establish strong baseline performances on this task and demonstrate that integration between visual and textual commonsense reasoning is the key and wins over non-integrative alternatives.
Multi-modal pretraining for learning high-level multi-modal representation is a further step towards deep learning and artificial intelligence. In this work, we propose a novel model, namely InterBERT (BERT for Interaction), which is the first model of our series of multimodal pretraining methods M6 (MultiModality-to-MultiModality Multitask Mega-transformer). The model owns strong capability of modeling interaction between the information flows of different modalities. The single-stream interaction module is capable of effectively processing information of multiple modalilties, and the two-stream module on top preserves the independence of each modality to avoid performance downgrade in single-modal tasks. We pretrain the model with three pretraining tasks, including masked segment modeling (MSM), masked region modeling (MRM) and image-text matching (ITM); and finetune the model on a series of vision-and-language downstream tasks. Experimental results demonstrate that InterBERT outperforms a series of strong baselines, including the most recent multi-modal pretraining methods, and the analysis shows that MSM and MRM are effective for pretraining and our method can achieve performances comparable to BERT in single-modal tasks. Besides, we propose a large-scale dataset for multi-modal pretraining in Chinese, and we develop the Chinese InterBERT which is the first Chinese multi-modal pretrained model. We pretrain the Chinese InterBERT on our proposed dataset of 3.1M image-text pairs from the mobile Taobao, the largest Chinese e-commerce platform. We finetune the model for text-based image retrieval, and recently we deployed the model online for topic-based recommendation.
Reasoning is an important ability that we learn from a very early age. Yet, reasoning is extremely hard for algorithms. Despite impressive recent progress that has been reported on tasks that necessitate reasoning, such as visual question answering and visual dialog, models often exploit biases in datasets. To develop models with better reasoning abilities, recently, the new visual commonsense reasoning (VCR) task has been introduced. Not only do models have to answer questions, but also do they have to provide a reason for the given answer. The proposed baseline achieved compelling results, leveraging a meticulously designed model composed of LSTM modules and attention nets. Here we show that a much simpler model obtained by ablating and pruning the existing intricate baseline can perform better with half the number of trainable parameters. By associating visual features with attribute information and better text to image grounding, we obtain further improvements for our simpler & effective baseline, TAB-VCR. We show that this approach results in a 5.3%, 4.4% and 6.5% absolute improvement over the previous state-of-the-art on question answering, answer justification and holistic VCR.
Visual commonsense reasoning task aims at leading the research field into solving cognition-level reasoning with the ability of predicting correct answers and meanwhile providing convincing reasoning paths, resulting in three sub-tasks i.e., Q->A, QA->R and Q->AR. It poses great challenges over the proper semantic alignment between vision and linguistic domains and knowledge reasoning to generate persuasive reasoning paths. Existing works either resort to a powerful end-to-end network that cannot produce interpretable reasoning paths or solely explore intra-relationship of visual objects (homogeneous graph) while ignoring the cross-domain semantic alignment among visual concepts and linguistic words. In this paper, we propose a new Heterogeneous Graph Learning (HGL) framework for seamlessly integrating the intra-graph and inter-graph reasoning in order to bridge vision and language domain. Our HGL consists of a primal vision-to-answer heterogeneous graph (VAHG) module and a dual question-to-answer heterogeneous graph (QAHG) module to interactively refine reasoning paths for semantic agreement. Moreover, our HGL integrates a contextual voting module to exploit a long-range visual context for better global reasoning. Experiments on the large-scale Visual Commonsense Reasoning benchmark demonstrate the superior performance of our proposed modules on three tasks (improving 5% accuracy on Q->A, 3.5% on QA->R, 5.8% on Q->AR)
The task of Visual Commonsense Reasoning is extremely challenging in the sense that the model has to not only be able to answer a question given an image, but also be able to learn to reason. The baselines introduced in this task are quite limiting because two networks are trained for predicting answers and rationales separately. Question and image is used as input to train answer prediction network while question, image and correct answer are used as input in the rationale prediction network. As rationale is conditioned on the correct answer, it is based on the assumption that we can solve Visual Question Answering task without any error - which is over ambitious. Moreover, such an approach makes both answer and rationale prediction two completely independent VQA tasks rendering cognition task meaningless. In this paper, we seek to address these issues by proposing an end-to-end trainable model which considers both answers and their reasons jointly. Specifically, we first predict the answer for the question and then use the chosen answer to predict the rationale. However, a trivial design of such a model becomes non-differentiable which makes it difficult to train. We solve this issue by proposing four approaches - softmax, gumbel-softmax, reinforcement learning based sampling and direct cross entropy against all pairs of answers and rationales. We demonstrate through experiments that our model performs competitively against current state-of-the-art. We conclude with an analysis of presented approaches and discuss avenues for further work.
Joint image-text embedding is the bedrock for most Vision-and-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 fine-grained alignment between words and image regions during pre-training. Comprehensive analysis shows that both conditional masking and OT-based 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 Answering, Image-Text Retrieval, Referring Expression Comprehension, Visual Commonsense Reasoning, Visual Entailment, and NLVR$^2$. Code is available at https://github.com/ChenRocks/UNITER.
We introduce a new pre-trainable generic representation for visual-linguistic tasks, called Visual-Linguistic BERT (VL-BERT for short). VL-BERT adopts the simple yet powerful Transformer model as the backbone, and extends it to take both visual and linguistic embedded features as input. In it, each element of the input is either of a word from the input sentence, or a region-of-interest (RoI) from the input image. It is designed to fit for most of the visual-linguistic downstream tasks. To better exploit the generic representation, we pre-train VL-BERT on the massive-scale Conceptual Captions dataset, together with text-only corpus. Extensive empirical analysis demonstrates that the pre-training procedure can better align the visual-linguistic clues and benefit the downstream tasks, such as visual commonsense reasoning, visual question answering and referring expression comprehension. It is worth noting that VL-BERT achieved the first place of single model on the leaderboard of the VCR benchmark. Code is released at \url{https://github.com/jackroos/VL-BERT}.
We propose Unicoder-VL, a universal encoder that aims to learn joint representations of vision and language in a pre-training manner. Borrow ideas from cross-lingual pre-trained models, such as XLM and Unicoder, both visual and linguistic contents are fed into a multi-layer Transformer for the cross-modal pre-training, where three pre-trained tasks are employed, including Masked Language Modeling (MLM), Masked Object Classification (MOC) and Visual-linguistic Matching (VLM). The first two tasks learn context-aware representations for input tokens based on linguistic and visual contents jointly. The last task tries to predict whether an image and a text describe each other. After pretraining on large-scale image-caption pairs, we transfer Unicoder-VL to caption-based image-text retrieval and visual commonsense reasoning, with just one additional output layer. We achieve state-of-the-art or comparable results on both two tasks and show the powerful ability of the cross-modal pre-training.
To advance models of multimodal context, we introduce a simple yet powerful neural architecture for data that combines vision and natural language. The "Bounding Boxes in Text Transformer" (B2T2) also leverages referential information binding words to portions of the image in a single unified architecture. B2T2 is highly effective on the Visual Commonsense Reasoning benchmark (https://visualcommonsense.com), achieving a new state-of-the-art with a 25% relative reduction in error rate compared to published baselines and obtaining the best performance to date on the public leaderboard (as of May 22, 2019). A detailed ablation analysis shows that the early integration of the visual features into the text analysis is key to the effectiveness of the new architecture. A reference implementation of our models is provided (https://github.com/google-research/language/tree/master/language/question_answering/b2t2).
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, pro-cessing 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 models -- achieving 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.
Visual understanding goes well beyond object recognition. With one glance at an image, we can effortlessly imagine the world beyond the pixels: for instance, we can infer people's actions, goals, and mental states. While this task is easy for humans, it is tremendously difficult for today's vision systems, requiring higher-order cognition and commonsense reasoning about the world. We formalize this task as Visual Commonsense Reasoning. Given a challenging question about an image, a machine must answer correctly and then provide a rationale justifying its answer. Next, we introduce a new dataset, VCR, consisting of 290k multiple choice QA problems derived from 110k movie scenes. The key recipe for generating non-trivial and high-quality problems at scale is Adversarial Matching, a new approach to transform rich annotations into multiple choice questions with minimal bias. Experimental results show that while humans find VCR easy (over 90% accuracy), state-of-the-art vision models struggle (~45%). To move towards cognition-level understanding, we present a new reasoning engine, Recognition to Cognition Networks (R2C), that models the necessary layered inferences for grounding, contextualization, and reasoning. R2C helps narrow the gap between humans and machines (~65%); still, the challenge is far from solved, and we provide analysis that suggests avenues for future work.