Masked language models (MLMs) are pretrained with a denoising objective that, while useful, is in a mismatch with the objective of downstream fine-tuning. We propose pragmatic masking and surrogate fine-tuning as two strategies that exploit social cues to drive pre-trained representations toward a broad set of concepts useful for a wide class of social meaning tasks. To test our methods, we introduce a new benchmark of 15 different Twitter datasets for social meaning detection. Our methods achieve 2.34% F1 over a competitive baseline, while outperforming other transfer learning methods such as multi-task learning and domain-specific language models pretrained on large datasets. With only 5% of training data (severely few-shot), our methods enable an impressive 68.74% average F1, and we observe promising results in a zero-shot setting involving six datasets from three different languages.

Stereo matching has recently witnessed remarkable progress using Deep Neural Networks (DNNs). But, how robust are they? Although it has been well-known that DNNs often suffer from adversarial vulnerability with a catastrophic drop in performance, the situation is even worse in stereo matching. This paper first shows that a type of weak white-box attacks can fail state-of-the-art methods. The attack is learned by a proposed stereo-constrained projected gradient descent (PGD) method in stereo matching. This observation raises serious concerns for the deployment of DNN-based stereo matching. Parallel to the adversarial vulnerability, DNN-based stereo matching is typically trained under the so-called simulation to reality pipeline, and thus domain generalizability is an important problem. This paper proposes to rethink the learnable DNN-based feature backbone towards adversarially-robust and domain generalizable stereo matching, either by completely removing it or by applying it only to the left reference image. It computes the matching cost volume using the classic multi-scale census transform (i.e., local binary pattern) of the raw input stereo images, followed by a stacked Hourglass head sub-network solving the matching problem. In experiments, the proposed method is tested in the SceneFlow dataset and the KITTI2015 benchmark. It significantly improves the adversarial robustness, while retaining accuracy performance comparable to state-of-the-art methods. It also shows better generalizability from simulation (SceneFlow) to real (KITTI) datasets when no fine-tuning is used.

Despite the progress in automatic detection of radiologic findings from chest X-ray (CXR) images in recent years, a quantitative evaluation of the explainability of these models is hampered by the lack of locally labeled datasets for different findings. With the exception of a few expert-labeled small-scale datasets for specific findings, such as pneumonia and pneumothorax, most of the CXR deep learning models to date are trained on global "weak" labels extracted from text reports, or trained via a joint image and unstructured text learning strategy. Inspired by the Visual Genome effort in the computer vision community, we constructed the first Chest ImaGenome dataset with a scene graph data structure to describe $242,072$ images. Local annotations are automatically produced using a joint rule-based natural language processing (NLP) and atlas-based bounding box detection pipeline. Through a radiologist constructed CXR ontology, the annotations for each CXR are connected as an anatomy-centered scene graph, useful for image-level reasoning and multimodal fusion applications. Overall, we provide: i) $1,256$ combinations of relation annotations between $29$ CXR anatomical locations (objects with bounding box coordinates) and their attributes, structured as a scene graph per image, ii) over $670,000$ localized comparison relations (for improved, worsened, or no change) between the anatomical locations across sequential exams, as well as ii) a manually annotated gold standard scene graph dataset from $500$ unique patients.

Dealing with missing values and incomplete time series is a labor-intensive and time-consuming inevitable task when handling data coming from real-world applications. Effective spatio-temporal representations would allow imputation methods to reconstruct missing temporal data by exploiting information coming from sensors at different locations. However, standard methods fall short in capturing the nonlinear time and space dependencies existing within networks of interconnected sensors and do not take full advantage of the available - and often strong - relational information. Notably, most of state-of-the-art imputation methods based on deep learning do not explicitly model relational aspects and, in any case, do not exploit processing frameworks able to adequately represent structured spatio-temporal data. Conversely, graph neural networks have recently surged in popularity as both expressive and scalable tools for processing sequential data with relational inductive biases. In this work, we present the first assessment of graph neural networks in the context of multivariate time series imputation. In particular, we introduce a novel graph neural network architecture, named GRIL, which aims at reconstructing missing data in the different channels of a multivariate time series by learning spatial-temporal representations through message passing. Preliminary empirical results show that our model outperforms state-of-the-art methods in the imputation task on relevant benchmarks with mean absolute error improvements often higher than 20%.

Opinion prediction is an emerging research area with diverse real-world applications, such as market research and situational awareness. We identify two lines of approaches to the problem of opinion prediction. One uses topic-based sentiment analysis with time-series modeling, while the other uses static embedding of text. The latter approaches seek user-specific solutions by generating user fingerprints. Such approaches are useful in predicting user's reactions to unseen content. In this work, we propose a novel dynamic fingerprinting method that leverages contextual embedding of user's comments conditioned on relevant user's reading history. We integrate BERT variants with a recurrent neural network to generate predictions. The results show up to 13\% improvement in micro F1-score compared to previous approaches. Experimental results show novel insights that were previously unknown such as better predictions for an increase in dynamic history length, the impact of the nature of the article on performance, thereby laying the foundation for further research.

The lottery ticket hypothesis (LTH) [19] claims that randomly-initialized, dense neural networks contain (sparse) subnetworks that, when trained an equal amount in isolation, can match the dense network's performance. Although LTH is useful for discovering efficient network architectures, its three-step process--pre-training, pruning, and re-training--is computationally expensive, as the dense model must be fully pre-trained. Luckily, "early-bird" tickets can be discovered within neural networks that are minimally pre-trained [67], allowing for the creation of efficient, LTH-inspired training procedures. Yet, no theoretical foundation of this phenomenon exists. We derive an analytical bound for the number of pre-training iterations that must be performed for a winning ticket to be discovered, thus providing a theoretical understanding of when and why such early-bird tickets exist. By adopting a greedy forward selection pruning strategy [65], we directly connect the pruned network's performance to the loss of the dense network from which it was derived, revealing a threshold in the number of pre-training iterations beyond which high-performing subnetworks are guaranteed to exist. We demonstrate the validity of our theoretical results across a variety of architectures and datasets, including multi-layer perceptrons (MLPs) trained on MNIST and several deep convolutional neural network (CNN) architectures trained on CIFAR10 and ImageNet.

Estimating the pose of animals can facilitate the understanding of animal motion which is fundamental in disciplines such as biomechanics, neuroscience, ethology, robotics and the entertainment industry. Human pose estimation models have achieved high performance due to the huge amount of training data available. Achieving the same results for animal pose estimation is challenging due to the lack of animal pose datasets. To address this problem we introduce SyDog: a synthetic dataset of dogs containing ground truth pose and bounding box coordinates which was generated using the game engine, Unity. We demonstrate that pose estimation models trained on SyDog achieve better performance than models trained purely on real data and significantly reduce the need for the labour intensive labelling of images. We release the SyDog dataset as a training and evaluation benchmark for research in animal motion.

Data selection methods, such as active learning and core-set selection, are useful tools for improving the data efficiency of deep learning models on large-scale datasets. However, recent deep learning models have moved forward from independent and identically distributed data to graph-structured data, such as social networks, e-commerce user-item graphs, and knowledge graphs. This evolution has led to the emergence of Graph Neural Networks (GNNs) that go beyond the models existing data selection methods are designed for. Therefore, we present Grain, an efficient framework that opens up a new perspective through connecting data selection in GNNs with social influence maximization. By exploiting the common patterns of GNNs, Grain introduces a novel feature propagation concept, a diversified influence maximization objective with novel influence and diversity functions, and a greedy algorithm with an approximation guarantee into a unified framework. Empirical studies on public datasets demonstrate that Grain significantly improves both the performance and efficiency of data selection (including active learning and core-set selection) for GNNs. To the best of our knowledge, this is the first attempt to bridge two largely parallel threads of research, data selection, and social influence maximization, in the setting of GNNs, paving new ways for improving data efficiency.

Current one-stage methods for visual grounding encode the language query as one holistic sentence embedding before fusion with visual feature. Such a formulation does not treat each word of a query sentence on par when modeling language to visual attention, therefore prone to neglect words which are less important for sentence embedding but critical for visual grounding. In this paper we propose Word2Pix: a one-stage visual grounding network based on encoder-decoder transformer architecture that enables learning for textual to visual feature correspondence via word to pixel attention. The embedding of each word from the query sentence is treated alike by attending to visual pixels individually instead of single holistic sentence embedding. In this way, each word is given equivalent opportunity to adjust the language to vision attention towards the referent target through multiple stacks of transformer decoder layers. We conduct the experiments on RefCOCO, RefCOCO+ and RefCOCOg datasets and the proposed Word2Pix outperforms existing one-stage methods by a notable margin. The results obtained also show that Word2Pix surpasses two-stage visual grounding models, while at the same time keeping the merits of one-stage paradigm namely end-to-end training and real-time inference speed intact.

Imagine telling your chatbot that your dog just died. Instead In this work, we introduce Knowledge-Embedded Attention of correctly understanding that you are experiencing grief (and (KEA), a knowledge-augmented attention mechanism that offering condolences), it classifies you as feeling sad and offers enriches the contextual representation provided by pre-trained to play you a happy song to cheer you up. People experience language models using emotional information obtained from a wide range of emotions, and it is important for AI agents external knowledge sources. This is achieved by incorporating to correctly recognize subtle differences between emotions the encoded emotional knowledge with the contextual representations like sadness and grief, in order to improve their interactions to form a modified key matrix. This key matrix with people and to avoid making a faux pas like the chatbot is then used to attend to the contextual representations to above [1]. Traditionally, the vast majority of work in emotion construct a more emotionally-aware representation of the input recognition from text focuses on recognizing just six "basic" text that can be used to recognise emotions. We introduce two emotions [2], [3], usually happiness, surprise, sadness, anger, variants of KEA, (i) a word-level KEA and (ii) a sentencelevel disgust, and fear. This set clearly fails to capture the broad KEA, which incorporate knowledge at different text spectrum of emotions that people experience and express in granularities.