This study uses TikTok (N = 8,173) to examine how short-form video platforms challenge the protest paradigm in the recent Black Lives Matter movement. A computer-mediated visual analysis, computer vision, is employed to identify the presence of four visual frames of protest (riot, confrontation, spectacle, and debate) in multimedia content. Results of descriptive statistics and the t-test indicate that the three delegitimizing frames - riot, confrontation, and spectacle - are rarely found on TikTok, whereas the debate frame, that empowers marginalized communities, dominates the public sphere. However, although the three delegitimizing frames receive lower social media visibility, as measured by views, likes, shares, followers, and durations, legitimizing elements, such as the debate frame, minority identities, and unofficial sources, are not generally favored by TikTok audiences. This study concludes that while short-form video platforms could potentially challenge the protest paradigm on the content creators' side, the audiences' preference as measured by social media visibility might still be moderately associated with the protest paradigm.
Transferring knowledge from one domain to another is of practical importance for many tasks in natural language processing, especially when the amount of available data in the target domain is limited. In this work, we propose a novel few-shot approach to domain adaptation in the context of Named Entity Recognition (NER). We propose a two-step approach consisting of a variable base module and a template module that leverages the knowledge captured in pre-trained language models with the help of simple descriptive patterns. Our approach is simple yet versatile and can be applied in few-shot and zero-shot settings. Evaluating our lightweight approach across a number of different datasets shows that it can boost the performance of state-of-the-art baselines by 2-5% F1-score.
As social robots and other intelligent machines enter the home, artificial emotional intelligence (AEI) is taking center stage to address users' desire for deeper, more meaningful human-machine interaction. To accomplish such efficacious interaction, the next-generation AEI need comprehensive human emotion models for training. Unlike theory of emotion, which has been the historical focus in psychology, emotion models are a descriptive tools. In practice, the strongest models need robust coverage, which means defining the smallest core set of emotions from which all others can be derived. To achieve the desired coverage, we turn to word embeddings from natural language processing. Using unsupervised clustering techniques, our experiments show that with as few as 15 discrete emotion categories, we can provide maximum coverage across six major languages--Arabic, Chinese, English, French, Spanish, and Russian. In support of our findings, we also examine annotations from two large-scale emotion recognition datasets to assess the validity of existing emotion models compared to human perception at scale. Because robust, comprehensive emotion models are foundational for developing real-world affective computing applications, this work has broad implications in social robotics, human-machine interaction, mental healthcare, and computational psychology.
The quest for seeking health information has swamped the web with consumers' health-related questions. Generally, consumers use overly descriptive and peripheral information to express their medical condition or other healthcare needs, contributing to the challenges of natural language understanding. One way to address this challenge is to summarize the questions and distill the key information of the original question. To address this issue, we introduce a new dataset, CHQ-Summ that contains 1507 domain-expert annotated consumer health questions and corresponding summaries. The dataset is derived from the community question-answering forum and therefore provides a valuable resource for understanding consumer health-related posts on social media. We benchmark the dataset on multiple state-of-the-art summarization models to show the effectiveness of the dataset.
In AI research, so far, the attention paid to the characterization and representation of function and affordance has been sporadic and sparse, even though this aspect features prominently in an intelligent system's functioning. In the sporadic and sparse, though commendable efforts so far devoted to the characterization and understanding of function and affordance, there has also been no general framework that could unify all the different use domains and situations related to the representation and application of functional concepts. This paper develops just such a general framework, with an approach that emphasizes the fact that the representations involved must be explicitly cognitive and conceptual, and they must also contain causal characterizations of the events and processes involved, as well as employ conceptual constructs that are grounded in the referents to which they refer, in order to achieve maximal generality. The basic general framework is described, along with a set of basic guiding principles with regards to the representation of functionality. To properly and adequately characterize and represent functionality, a descriptive representation language is needed. This language is defined and developed, and many examples of its use are described. The general framework is developed based on an extension of the general language meaning representational framework called conceptual dependency. To support the general characterization and representation of functionality, the basic conceptual dependency framework is enhanced with representational devices called structure anchor and conceptual dependency elaboration, together with the definition of a set of ground level concepts. These novel representational constructs are defined, developed, and described. A general framework dealing with functionality would represent a major step toward achieving Artificial General Intelligence.
Discrimination in machine learning often arises along multiple dimensions (a.k.a. protected attributes); it is then desirable to ensure \emph{intersectional fairness} -- i.e., that no subgroup is discriminated against. It is known that ensuring \emph{marginal fairness} for every dimension independently is not sufficient in general. Due to the exponential number of subgroups, however, directly measuring intersectional fairness from data is impossible. In this paper, our primary goal is to understand in detail the relationship between marginal and intersectional fairness through statistical analysis. We first identify a set of sufficient conditions under which an exact relationship can be obtained. Then, we prove bounds (easily computable through marginal fairness and other meaningful statistical quantities) in high-probability on intersectional fairness in the general case. Beyond their descriptive value, we show that these theoretical bounds can be leveraged to derive a heuristic improving the approximation and bounds of intersectional fairness by choosing, in a relevant manner, protected attributes for which we describe intersectional subgroups. Finally, we test the performance of our approximations and bounds on real and synthetic data-sets.
In recent years, the field of neuroscience has gone through rapid experimental advances and a significant increase in the use of quantitative and computational methods. This growth has created a need for clearer analyses of the theory and modeling approaches used in the field. This issue is particularly complex in neuroscience because the field studies phenomena across a wide range of scales and often requires consideration of these phenomena at varying degrees of abstraction, from precise biophysical interactions to the computations they implement. We argue that a pragmatic perspective of science, in which descriptive, mechanistic, and normative approaches each play a distinct role in defining and bridging levels of abstraction will facilitate neuroscientific practice. This analysis leads to methodological suggestions, including selecting a level of abstraction that is appropriate for a given problem, identifying transfer functions to connect models and data, and the use of models themselves as a form of experiment.
The use of cameras and computational algorithms for noninvasive, low-cost and scalable measurement of physiological (e.g., cardiac and pulmonary) vital signs is very attractive. However, diverse data representing a range of environments, body motions, illumination conditions and physiological states is laborious, time consuming and expensive to obtain. Synthetic data have proven a valuable tool in several areas of machine learning, yet are not widely available for camera measurement of physiological states. Synthetic data offer "perfect" labels (e.g., without noise and with precise synchronization), labels that may not be possible to obtain otherwise (e.g., precise pixel level segmentation maps) and provide a high degree of control over variation and diversity in the dataset. We present SCAMPS, a dataset of synthetics containing 2,800 videos (1.68M frames) with aligned cardiac and respiratory signals and facial action intensities. The RGB frames are provided alongside segmentation maps. We provide precise descriptive statistics about the underlying waveforms, including inter-beat interval, heart rate variability, and pulse arrival time. Finally, we present baseline results training on these synthetic data and testing on real-world datasets to illustrate generalizability.
Metaverse is a virtual universe that combines the physical world and the digital world. People can socialize, play games and even shop with their avatars created in this virtual environment. Metaverse, which is growing very fast in terms of investment, is both a profitable and risky area for consumers. In order to enter the Metaverse for investment purposes, it is necessary to do a certain research and gather information. In this direction, the aim of the study is to determine the effect of the quality of the information obtained by the consumers about the metaverse world, the reliability of the information and the perceived risk, on the purchase intention from the point of view of the information adoption model. For the research, data were collected online from 495 consumers who were interested in metaverse investment. AMOS and SPSS package programs were used in the analysis. First, descriptive statistical analyzes were made for the basic structure of the variables. Then the reliability and validity of the model were tested. Finally, the structural equation model was used to test the proposed model. According to the findings, the reliability and quality of the information affect the purchase intention positively and significantly, while the perceived risk affects the purchase intention negatively and significantly.
Item ranking systems support users in multi-criteria decision-making tasks. Users need to trust rankings and ranking algorithms to reflect user preferences nicely while avoiding systematic errors and biases. However, today only few approaches help end users, model developers, and analysts to explain rankings. We report on the study of explanation approaches from the perspectives of recommender systems, explainable AI, and visualization research and propose the first cross-domain design space for explainers of item rankings. In addition, we leverage the descriptive power of the design space to characterize a) existing explainers and b) three main user groups involved in ranking explanation tasks. The generative power of the design space is a means for future designers and developers to create more target-oriented solutions in this only weakly exploited space.
Video understanding has achieved great success in representation learning, such as video caption, video object grounding, and video descriptive question-answer. However, current methods still struggle on video reasoning, including evidence reasoning and commonsense reasoning. To facilitate deeper video understanding towards video reasoning, we present the task of Causal-VidQA, which includes four types of questions ranging from scene description (description) to evidence reasoning (explanation) and commonsense reasoning (prediction and counterfactual). For commonsense reasoning, we set up a two-step solution by answering the question and providing a proper reason. Through extensive experiments on existing VideoQA methods, we find that the state-of-the-art methods are strong in descriptions but weak in reasoning. We hope that Causal-VidQA can guide the research of video understanding from representation learning to deeper reasoning. The dataset and related resources are available at \url{https://github.com/bcmi/Causal-VidQA.git}.
Many ontologies, i.e., Description Logic (DL) knowledge bases, have been developed to provide rich knowledge about various domains. An ontology consists of an ABox, i.e., assertion axioms between two entities or between a concept and an entity, and a TBox, i.e., terminology axioms between two concepts. Neural logical reasoning (NLR) is a fundamental task to explore such knowledge bases, which aims at answering multi-hop queries with logical operations based on distributed representations of queries and answers. While previous NLR methods can give specific entity-level answers, i.e., ABox answers, they are not able to provide descriptive concept-level answers, i.e., TBox answers, where each concept is a description of a set of entities. In other words, previous NLR methods only reason over the ABox of an ontology while ignoring the TBox. In particular, providing TBox answers enables inferring the explanations of each query with descriptive concepts, which make answers comprehensible to users and are of great usefulness in the field of applied ontology. In this work, we formulate the problem of neural logical reasoning across TBox and ABox (TA-NLR), solving which needs to address challenges in incorporating, representing, and operating on concepts. We propose an original solution named TAR for TA-NLR. Firstly, we incorporate description logic based ontological axioms to provide the source of concepts. Then, we represent concepts and queries as fuzzy sets, i.e., sets whose elements have degrees of membership, to bridge concepts and queries with entities. Moreover, we design operators involving concepts on top of fuzzy set representation of concepts and queries for optimization and inference. Extensive experimental results on two real-world datasets demonstrate the effectiveness of TAR for TA-NLR.
Modern image captioning models are usually trained with text similarity objectives. However, since reference captions in public datasets often describe the most salient common objects, models trained with text similarity objectives tend to ignore specific and detailed aspects of an image that distinguish it from others. Toward more descriptive and distinctive caption generation, we propose using CLIP, a multimodal encoder trained on huge image-text pairs from web, to calculate multimodal similarity and use it as a reward function. We also propose a simple finetuning strategy of the CLIP text encoder to improve grammar that does not require extra text annotation. This completely eliminates the need for reference captions during the reward computation. To comprehensively evaluate descriptive captions, we introduce FineCapEval, a new dataset for caption evaluation with fine-grained criteria: overall, background, object, relations. In our experiments on text-to-image retrieval and FineCapEval, the proposed CLIP-guided model generates more distinctive captions than the CIDEr-optimized model. We also show that our unsupervised grammar finetuning of the CLIP text encoder alleviates the degeneration problem of the naive CLIP reward. Lastly, we show human analysis where the annotators strongly prefer the CLIP reward to the CIDEr and MLE objectives according to various criteria. Code and Data: https://github.com/j-min/CLIP-Caption-Reward
Combining the visual modality with pretrained language models has been surprisingly effective for simple descriptive tasks such as image captioning. More general text generation however remains elusive. We take a step back and ask: How do these models work for more complex generative tasks, i.e. conditioning on both text and images? Are multimodal models simply visually adapted language models, or do they combine they reason jointly over modalities? We investigate these questions in the context of self-rationalization (jointly generating task labels/answers and free-text explanations) of three tasks: (i) visual question answering in VQA-X, (ii) visual commonsense reasoning in VCR, and (iii) visual-textual entailment in e-SNLI-VE. We show that recent unimodal advances, CLIP image representations and scaling of language models, do not consistently improve self-rationalization in multimodal tasks. We find that no single model type works universally best across tasks, datasets, and finetuning data sizes. Our findings motivate the need for novel general backbones approach that move text generation from images and text beyond image captioning.
Video captioning is a challenging task as it needs to accurately transform visual understanding into natural language description. To date, state-of-the-art methods inadequately model global-local representation across video frames for caption generation, leaving plenty of room for improvement. In this work, we approach the video captioning task from a new perspective and propose a GL-RG framework for video captioning, namely a \textbf{G}lobal-\textbf{L}ocal \textbf{R}epresentation \textbf{G}ranularity. Our GL-RG demonstrates three advantages over the prior efforts: 1) we explicitly exploit extensive visual representations from different video ranges to improve linguistic expression; 2) we devise a novel global-local encoder to produce rich semantic vocabulary to obtain a descriptive granularity of video contents across frames; 3) we develop an incremental training strategy which organizes model learning in an incremental fashion to incur an optimal captioning behavior. Experimental results on the challenging MSR-VTT and MSVD datasets show that our DL-RG outperforms recent state-of-the-art methods by a significant margin. Code is available at \url{https://github.com/ylqi/GL-RG}.
We propose DEER (Descriptive Knowledge Graph for Explaining Entity Relationships) - an open and informative form of modeling entity relationships. In DEER, relationships between entities are represented by free-text relation descriptions. For instance, the relationship between entities of machine learning and algorithm can be represented as ``Machine learning explores the study and construction of algorithms that can learn from and make predictions on data.'' To construct DEER, we propose a self-supervised learning method to extract relation descriptions with the analysis of dependency patterns and generate relation descriptions with a transformer-based relation description synthesizing model, where no human labeling is required. Experiments demonstrate that our system can extract and generate high-quality relation descriptions for explaining entity relationships. The results suggest that we can build an open and informative knowledge graph without human annotation.
What is the relation between a word and its description, or a word and its embedding? Both descriptions and embeddings are semantic representations of words. But, what information from the original word remains in these representations? Or more importantly, which information about a word do these two representations share? Definition Modeling and Reverse Dictionary are two opposite learning tasks that address these questions. The goal of the Definition Modeling task is to investigate the power of information laying inside a word embedding to express the meaning of the word in a humanly understandable way -- as a dictionary definition. Conversely, the Reverse Dictionary task explores the ability to predict word embeddings directly from its definition. In this paper, by tackling these two tasks, we are exploring the relationship between words and their semantic representations. We present our findings based on the descriptive, exploratory, and predictive data analysis conducted on the CODWOE dataset. We give a detailed overview of the systems that we designed for Definition Modeling and Reverse Dictionary tasks, and that achieved top scores on SemEval-2022 CODWOE challenge in several subtasks. We hope that our experimental results concerning the predictive models and the data analyses we provide will prove useful in future explorations of word representations and their relationships.
Blockchain introduces decentralized trust in peer-to-peer networks, advancing security and democratizing systems. Yet, a unified definition for decentralization remains elusive. Our Systematization of Knowledge (SoK) seeks to bridge this gap, emphasizing quantification and methodological coherence. We've formulated a taxonomy defining blockchain decentralization across five facets: consensus, network, governance, wealth, and transaction. Despite the prevalent focus on consensus decentralization, our novel index, based on Shannon entropy, provides comprehensive insights. Moreover, we delve into alternative metrics like the Gini and Nakamoto Coefficients and the Herfindahl-Hirschman Index (HHI), supplemented by an open-source Python tool on GitHub.In terms of methodology, blockchain research has often bypassed stringent scientific methods. By employing descriptive, predictive, and causal methods, our study showcases the potential of structured research in blockchain. Descriptively, we observe a trend of converging decentralization levels over time. Examining DeFi platforms reveals exchange and lending applications as more decentralized than their payment and derivatives counterparts. Predictively, there's a notable correlation between Ether's returns and transaction decentralization in Ether-backed stablecoins. Causally, Ethereum's transition to the EIP-1559 transaction fee model has a profound impact on DeFi transaction decentralization. To conclude, our work outlines directions for blockchain research, emphasizing the delicate balance among decentralization facets, fostering long-term decentralization, and the ties between decentralization, security, privacy, and efficiency. We end by spotlighting challenges in grasping blockchain decentralization intricacies.
Video captioning aims to understand the spatio-temporal semantic concept of the video and generate descriptive sentences. The de-facto approach to this task dictates a text generator to learn from \textit{offline-extracted} motion or appearance features from \textit{pre-trained} vision models. However, these methods may suffer from the so-called \textbf{\textit{"couple"}} drawbacks on both \textit{video spatio-temporal representation} and \textit{sentence generation}. For the former, \textbf{\textit{"couple"}} means learning spatio-temporal representation in a single model(3DCNN), resulting the problems named \emph{disconnection in task/pre-train domain} and \emph{hard for end-to-end training}. As for the latter, \textbf{\textit{"couple"}} means treating the generation of visual semantic and syntax-related words equally. To this end, we present $\mathcal{D}^{2}$ - a dual-level decoupled transformer pipeline to solve the above drawbacks: \emph{(i)} for video spatio-temporal representation, we decouple the process of it into "first-spatial-then-temporal" paradigm, releasing the potential of using dedicated model(\textit{e.g.} image-text pre-training) to connect the pre-training and downstream tasks, and makes the entire model end-to-end trainable. \emph{(ii)} for sentence generation, we propose \emph{Syntax-Aware Decoder} to dynamically measure the contribution of visual semantic and syntax-related words. Extensive experiments on three widely-used benchmarks (MSVD, MSR-VTT and VATEX) have shown great potential of the proposed $\mathcal{D}^{2}$ and surpassed the previous methods by a large margin in the task of video captioning.
Knowledge Distillation (KD) is a strategy for the definition of a set of transferability gangways to improve the efficiency of Convolutional Neural Networks. Feature-based Knowledge Distillation is a subfield of KD that relies on intermediate network representations, either unaltered or depth-reduced via maximum activation maps, as the source knowledge. In this paper, we propose and analyse the use of a 2D frequency transform of the activation maps before transferring them. We pose that\textemdash by using global image cues rather than pixel estimates, this strategy enhances knowledge transferability in tasks such as scene recognition, defined by strong spatial and contextual relationships between multiple and varied concepts. To validate the proposed method, an extensive evaluation of the state-of-the-art in scene recognition is presented. Experimental results provide strong evidences that the proposed strategy enables the student network to better focus on the relevant image areas learnt by the teacher network, hence leading to better descriptive features and higher transferred performance than every other state-of-the-art alternative. We publicly release the training and evaluation framework used along this paper at http://www-vpu.eps.uam.es/publications/DCTBasedKDForSceneRecognition.