Data dashboards are designed to help users manage data collected about them. However, prior work showed that exposure to some dashboards, notably Google's My Activity dashboard, results in significant decreases in perceived concern and increases in perceived benefit from data collection, contrary to expectations. We theorize that this result is due to the fact that data dashboards currently do not sufficiently "connect the dots" of the data food chain, that is, by connecting data collection with the use of that data. To evaluate this, we designed a study where participants assigned advertising interest labels to their own real activities, effectively acting as a behavioral advertising engine to "connect the dots." When comparing pre- and post-labeling task responses, we find no significant difference in concern with Google's data collection practices, which indicates that participants' priors are maintained after more exposure to the data food chain (differing from prior work), suggesting that data dashboards that offer deeper perspectives of how data collection is used have potential. However, these gains are offset when participants are exposed to their true interest labels inferred by Google. Concern for data collection dropped significantly as participants viewed Google's labeling as generic compared to their own more specific labeling. This presents a possible new paradox that must be overcome when designing data dashboards, the generic paradox, which occurs when users misalign individual, generic inferences from collected data as benign compared to the totality and specificity of many generic inferences made about them.
相關內容
Thoughtfully designing services and rigorously testing software to support personal information management (PIM) requires understanding the relevant collections, but relatively little is known about what people keep in their file collections, especially personal collections. Complementing recent work on the structure of 348 file collections, we examine those collections' contents, how much content is duplicated, and how collections used for personal matters differ from those used for study and work. Though all collections contain many images, some intuitively common file types are surprisingly scarce. Personal collections contain more audio than others, knowledge workers' collections contain more text documents but far fewer folders, and IT collections exhibit unusual traits. Collection duplication is correlated to collections' structural traits, but surprisingly, not to collection age. We discuss our findings in light of prior works and provide implications for various kinds of information research.
Large language models (LLMs) have played a pivotal role in building communicative AI to imitate human behaviors but face the challenge of efficient customization. To tackle this challenge, recent studies have delved into the realm of model editing, which manipulates specific memories of language models and changes the related language generation. However, the robustness of model editing remains an open question. This work seeks to understand the strengths and limitations of editing methods, thus facilitating robust, realistic applications of communicative AI. Concretely, we conduct extensive analysis to address the three key research questions. Q1: Can edited LLMs behave consistently resembling communicative AI in realistic situations? Q2: To what extent does the rephrasing of prompts lead LLMs to deviate from the edited knowledge memory? Q3: Which knowledge features are correlated with the performance and robustness of editing? Our experimental results uncover a substantial disparity between existing editing methods and the practical application of LLMs. On rephrased prompts that are complex and flexible but common in realistic applications, the performance of editing experiences a significant decline. Further analysis shows that more popular knowledge is memorized better, easier to recall, and more challenging to edit effectively.
In this work, we investigate to use Large Language Models (LLMs) for rumor detection on social media. However, it is challenging for LLMs to reason over the entire propagation information on social media, which contains news contents and numerous comments, due to LLMs may not concentrate on key clues in the complex propagation information, and have trouble in reasoning when facing massive and redundant information. Accordingly, we propose an LLM-empowered Rumor Detection (LeRuD) approach, in which we design prompts to teach LLMs to reason over important clues in news and comments, and divide the entire propagation information into a Chain-of-Propagation for reducing LLMs' burden. We conduct extensive experiments on the Twitter and Weibo datasets, and LeRuD outperforms several state-of-the-art rumor detection models by 3.2% to 7.7%. Meanwhile, by applying LLMs, LeRuD requires no data for training, and thus shows more promising rumor detection ability in few-shot or zero-shot scenarios.
This letter considers a two-panel massive multiple-input multiple-output (MIMO) communication system, where the base station (BS) is equipped with two antenna panels that may use different frequency bands for communication. By exploiting the geometric relationships between antenna panels, efficient channel inference methods across antenna panels are proposed to reduce the overhead of real-time channel estimation. Four scenarios are considered, namely far-field free-space, near-field free-space, multi-path sharing far-field scatterers, and multi-path sharing near-field scatterers. For both far-field and near-field free-space scenarios, we show that the channel of one panel can be fully inferred from that of the other panel, as long as the multi-path components (MPCs) composing the channel can be resolved. On the other hand, for the multi-path scenarios sharing far-field or near-field scatterers, only the angles or range of angles of the MPCs can be inferred, respectively. Simulation results based on commercial 3D ray-tracing software are presented to validate our developed channel inference techniques.
Large Language Models (LLMs) create exciting possibilities for powerful language processing tools to accelerate research in materials science. While LLMs have great potential to accelerate materials understanding and discovery, they currently fall short in being practical materials science tools. In this position paper, we show relevant failure cases of LLMs in materials science that reveal current limitations of LLMs related to comprehending and reasoning over complex, interconnected materials science knowledge. Given those shortcomings, we outline a framework for developing Materials Science LLMs (MatSci-LLMs) that are grounded in materials science knowledge and hypothesis generation followed by hypothesis testing. The path to attaining performant MatSci-LLMs rests in large part on building high-quality, multi-modal datasets sourced from scientific literature where various information extraction challenges persist. As such, we describe key materials science information extraction challenges which need to be overcome in order to build large-scale, multi-modal datasets that capture valuable materials science knowledge. Finally, we outline a roadmap for applying future MatSci-LLMs for real-world materials discovery via: 1. Automated Knowledge Base Generation; 2. Automated In-Silico Material Design; and 3. MatSci-LLM Integrated Self-Driving Materials Laboratories.
Several works related to spatial crowdsourcing have been proposed in the direction where the task executers are to perform the tasks within the stipulated deadlines. Though the deadlines are set, it may be a practical scenario that majority of the task executers submit the tasks as late as possible. This situation where the task executers may delay their task submission is termed as procrastination in behavioural economics. In many applications, these late submission of tasks may be problematic for task providers. So here, the participating agents (both task providers and task executers) are articulated with the procrastination issue. In literature, how to prevent this procrastination within the deadline is not addressed in spatial crowdsourcing scenario. However, in a bipartite graph setting one procrastination aware scheduling is proposed but balanced job (task and job will synonymously be used) distribution in different slots (also termed as schedules) is not considered there. In this paper, a procrastination aware scheduling of jobs is proliferated by proposing an (randomized) algorithm in spatial crowdsourcing scenario. Our algorithm ensures that balancing of jobs in different schedules are maintained. Our scheme is compared with the existing algorithm through extensive simulation and in terms of balancing effect, our proposed algorithm outperforms the existing one. Analytically it is shown that our proposed algorithm maintains the balanced distribution.
Deep Generative Models (DGMs) have been shown to be powerful tools for generating tabular data, as they have been increasingly able to capture the complex distributions that characterize them. However, to generate realistic synthetic data, it is often not enough to have a good approximation of their distribution, as it also requires compliance with constraints that encode essential background knowledge on the problem at hand. In this paper, we address this limitation and show how DGMs for tabular data can be transformed into Constrained Deep Generative Models (C-DGMs), whose generated samples are guaranteed to be compliant with the given constraints. This is achieved by automatically parsing the constraints and transforming them into a Constraint Layer (CL) seamlessly integrated with the DGM. Our extensive experimental analysis with various DGMs and tasks reveals that standard DGMs often violate constraints, some exceeding $95\%$ non-compliance, while their corresponding C-DGMs are never non-compliant. Then, we quantitatively demonstrate that, at training time, C-DGMs are able to exploit the background knowledge expressed by the constraints to outperform their standard counterparts with up to $6.5\%$ improvement in utility and detection. Further, we show how our CL does not necessarily need to be integrated at training time, as it can be also used as a guardrail at inference time, still producing some improvements in the overall performance of the models. Finally, we show that our CL does not hinder the sample generation time of the models.
Large Language Model (LLM) agents have been increasingly adopted as simulation tools to model humans in applications such as social science. However, one fundamental question remains: can LLM agents really simulate human behaviors? In this paper, we focus on one of the most critical behaviors in human interactions, trust, and aim to investigate whether or not LLM agents can simulate human trust behaviors. We first find that LLM agents generally exhibit trust behaviors, referred to as agent trust, under the framework of Trust Games, which are widely recognized in behavioral economics. Then, we discover that LLM agents can have high behavioral alignment with humans regarding trust behaviors, indicating the feasibility to simulate human trust behaviors with LLM agents. In addition, we probe into the biases in agent trust and the differences in agent trust towards agents and humans. We also explore the intrinsic properties of agent trust under conditions including advanced reasoning strategies and external manipulations. We further offer important implications for various scenarios where trust is paramount. Our study represents a significant step in understanding the behaviors of LLM agents and the LLM-human analogy.
This paper shows that masked autoencoders (MAE) are scalable self-supervised learners for computer vision. Our MAE approach is simple: we mask random patches of the input image and reconstruct the missing pixels. It is based on two core designs. First, we develop an asymmetric encoder-decoder architecture, with an encoder that operates only on the visible subset of patches (without mask tokens), along with a lightweight decoder that reconstructs the original image from the latent representation and mask tokens. Second, we find that masking a high proportion of the input image, e.g., 75%, yields a nontrivial and meaningful self-supervisory task. Coupling these two designs enables us to train large models efficiently and effectively: we accelerate training (by 3x or more) and improve accuracy. Our scalable approach allows for learning high-capacity models that generalize well: e.g., a vanilla ViT-Huge model achieves the best accuracy (87.8%) among methods that use only ImageNet-1K data. Transfer performance in downstream tasks outperforms supervised pre-training and shows promising scaling behavior.
Graph Neural Networks (GNNs) have recently been used for node and graph classification tasks with great success, but GNNs model dependencies among the attributes of nearby neighboring nodes rather than dependencies among observed node labels. In this work, we consider the task of inductive node classification using GNNs in supervised and semi-supervised settings, with the goal of incorporating label dependencies. Because current GNNs are not universal (i.e., most-expressive) graph representations, we propose a general collective learning approach to increase the representation power of any existing GNN. Our framework combines ideas from collective classification with self-supervised learning, and uses a Monte Carlo approach to sampling embeddings for inductive learning across graphs. We evaluate performance on five real-world network datasets and demonstrate consistent, significant improvement in node classification accuracy, for a variety of state-of-the-art GNNs.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191