Since the explosion in popularity of ChatGPT, large language models (LLMs) have continued to impact our everyday lives. Equipped with external tools that are designed for a specific purpose (e.g., for flight booking or an alarm clock), LLM agents exercise an increasing capability to assist humans in their daily work. Although LLM agents have shown a promising blueprint as daily assistants, there is a limited understanding of how they can provide daily assistance based on planning and sequential decision making capabilities. We draw inspiration from recent work that has highlighted the value of g'LLM-modulo' setups in conjunction with humans-in-the-loop for planning tasks. We conducted an empirical study (N = 248) of LLM agents as daily assistants in six commonly occurring tasks with different levels of risk typically associated with them (e.g., flight ticket booking and credit card payments). To ensure user agency and control over the LLM agent, we adopted LLM agents in a plan-then-execute manner, wherein the agents conducted step-wise planning and step-by-step execution in a simulation environment. We analyzed how user involvement at each stage affects their trust and collaborative team performance. Our findings demonstrate that LLM agents can be a double-edged sword - (1) they can work well when a high-quality plan and necessary user involvement in execution are available, and (2) users can easily mistrust the LLM agents with plans that seem plausible. We synthesized key insights for using LLM agents as daily assistants to calibrate user trust and achieve better overall task outcomes. Our work has important implications for the future design of daily assistants and human-AI collaboration with LLM agents.

This Special Issue of the Journal of Data and Information Quality (JDIQ) contains novel theoretical and methodological contributions on data curation involving humans in the loop. In this editorial, we summarize the scope of the issue and briefly describe its content.

Although data quality is a long-standing and enduring problem, it has recently received a resurgence of attention due to the fast proliferation of data analytics, machine learning, and decision-support applications built upon the wide-scale availability and accessibility of (big) data. The success of such applications heavily relies on not only the quantity, but also the quality of data. Data curation, which may include annotation, cleaning, transformation, integration, etc., is a critical step to provide adequate assurances on the quality of analytics and machine learning results. Such data preparation activities are recognised as time and resource intensive for data scientists as data often comes with a number of challenges that need to be tackled before it can be used in practice. Data re-purposing and the resulting distance between design and use intentions of the data, is a fundamental issue behind many of these challenges. These challenges include a variety of data issues such as noise and outliers, incompleteness, representativeness or biases, heterogeneity of format or semantics, etc. Mishandling these challenges can lead to negative and sometimes damaging effects, especially in critical domains like healthcare, transport, and finance. An observable distinct feature of data quality in these contexts is the increasingly important role played by humans, being often the source of data generation and the active players in data curation. This workshop will provide an opportunity to explore the interdisciplinary overlap between manual, automated, and hybrid human-machine methods of data curation.

Crowdsourcing is a popular technique to collect large amounts of human-generated labels, such as relevance judgments used to create information retrieval (IR) evaluation collections. Previous research has shown how collecting high quality labels from a crowdsourcing platform can be challenging. Existing quality assurance techniques focus on answer aggregation or on the use of gold questions where ground-truth data allows to check for the quality of the responses. In this paper, we present qualitative and quantitative results, revealing how different crowd workers adopt different work strate- gies to complete relevance judgment tasks efficiently and their consequent impact on quality. We delve into the techniques and tools that highly experienced crowd workers use to be more effi- cient in completing crowdsourcing micro-tasks. To this end, we use both qualitative results from worker interviews and surveys, as well as the results of a data-driven study of behavioral log data (i.e., clicks, keystrokes and keyboard shortcuts) collected from crowd workers performing relevance judgment tasks. Our results high- light the presence of frequently used shortcut patterns that can speed-up task completion, thus increasing the hourly wage of effi- cient workers. We observe how crowd work experiences result in different types of working strategies, productivity levels, quality and diversity of the crowdsourced judgments.

Paid micro-task crowdsourcing has gained in popularity partly due to the increasing need for large-scale manually labelled datasets which are often used to train and evaluate Artificial Intelligence systems. Modern paid crowdsourcing platforms use a piecework approach to rewards, meaning that workers are paid for each task they complete, given that their work quality is considered sufficient by the requester or the platform. Such an approach creates risks for workers; their work may be rejected without being rewarded, and they may be working on poorly rewarded tasks, in light of the disproportionate time required to complete them. As a result, recent research has shown that crowd workers may tend to choose specific, simple, and familiar tasks and avoid new requesters to manage these risks.
In this paper, we propose a novel crowdsourcing reward mechanism that allows workers to share these risks and achieve a standardized hourly wage equal for all participating workers. Reward-focused workers can thereby take up challenging and complex HITs without bearing the financial risk of not being rewarded for completed work. We experimentally compare different crowd reward schemes and observe their impact on worker performance and satisfaction. Our results show that 1) workers clearly perceive the benefits of the proposed reward scheme, 2) work effectiveness and efficiency are not impacted as compared to those of the piecework scheme, and 3) the presence of slow workers is limited and does not disrupt the proposed cooperation-based approaches.

This paper presents Scalpel-CD, a first-of-its-kind system that leverages both human and machine intelligence to debug noisy labels from the training data of machine learning systems. Our system identifies potentially wrong labels using a deep probabilistic model, which is able to infer the latent class of a high-dimensional data instance by exploiting data distributions in the underlying latent feature space. To minimize crowd efforts, it employs a data sampler which selects data instances that would benefit the most from being inspected by the crowd. The manually verified labels are then propagated to similar data instances in the original training data by exploiting the underlying data structure, thus scaling out the contribution from the crowd. Scalpel-CD is designed with a set of algorithmic solutions to automatically search for the optimal configurations for different types of training data, in terms of the underlying data structure, noise ratio, and noise types (random vs. structural). In a real deployment on multiple machine learning tasks, we demonstrate that Scalpel-CD is able to improve label quality by 12.9% with only 2.8% instances inspected by the crowd.

Complexity is crucial to characterize tasks performed by humans through computer systems. Yet, the theory and practice of crowdsourcing currently lacks a clear understanding of task complexity, hindering the design of effective and efficient execution interfaces or fair monetary rewards. To understand how complexity is perceived and distributed over crowdsourcing tasks, we instrumented an experiment where we asked workers to evaluate the complexity of 61 real-world re-instantiated crowdsourcing tasks. We show that task complexity, while being subjective, is coherently perceived across workers; on the other hand, it is significantly influenced by task type. Next, we develop a high-dimensional regression model, to assess the influence of three classes of structural features (metadata, content, and visual) on task complexity, and ultimately use them to measure task complexity. Results show that both the appearance and the language used in task description can accurately predict task complexity. Finally, we apply the same feature set to predict task performance, based on a set of 5 years-worth tasks in Amazon MTurk. Results show that features related to task complexity can improve the quality of task performance prediction, thus demonstrating the utility of complexity as a task modeling property.