In the big data era, it's important to identify trustworthy information from an influx of noisy contributed by unvetted sources online social media (e.g., Twitter, Instagram, Flickr). This task is referred as truth discovery which aims at identifying reliability and truthfulness claims they make without knowing either them a priori. There are two challenges that have not been well addressed in current solutions. The first one "misinformation spread" where majority contributing false claims,...

With the rapid growth of online social media and ubiquitous Internet connectivity, sensing has emerged as a new crowdsourcing application paradigm collecting observations (often called claims) about physical environment from humans or devices on their behalf. A fundamental problem in applications lies effectively ascertaining correctness claims reliability data sources without knowing either them priori, which is referred to truth discovery. While significant progress been made solve...

This work is motivated by the emergence of social sensing as a new paradigm collecting observations about physical environment from humans or devices on their behalf. These may be true false, and hence are viewed binary claims. A fundamental problem in applications lies ascertaining correctness claims reliability data sources without knowing either them priori. We refer to this truth discovery. Prior works have made significant progress addressing discovery problem, but two limitations...

This paper presents a confidence-aware maximum likelihood estimation framework to solve the truth problem in social sensing applications. Social has emerged as new paradigm of data collection, where group individuals volunteer (or are recruited) share certain observations or measurements about physical world. A key challenge applications lies ascertaining correctness reported from unvetted sources with unknown reliability. We refer this estimation. The prior works have made significant...

This work is motivated by the emergence of social sensing as a new paradigm collecting observations about physical environment from humans or devices on their behalf. These may be true false, and hence are viewed binary claims. A fundamental problem in applications lies ascertaining correctness claims reliability data sources without knowing either them priori. We refer to this truth discovery. Prior works have made significant progress addressing discovery problem, but two limitations...

Mobile crowdsourcing platforms often want to incentivize workers finish tasks with high quality and truthfully report their solutions by providing proper rewards. Most existing incentive mechanisms reward based on the comparison among workers' reported solutions. However, these are vulnerable worker collusion, i.e., coordinate misreport We address such an issue proposing a novel rewarding mechanism <inline-formula><tex-math notation="LaTeX">${truth detection}$</tex-math></inline-formula>...

Learning effective geospatial embeddings is crucial for a series of applications such as city analytics and earth monitoring. However, learning comprehensive region representations presents two significant challenges: first, the deficiency intra-region feature representation; second, difficulty from intricate inter-region dependencies. In this paper, we present GeoHG, an heterogeneous graph structure various downstream tasks. Specifically, tailor satellite image representation through...

Social sensing is a new application paradigm of cyber-physical-social systems (CPSS), where group individuals volunteer to report their claims about the physical environment using cyber devices. A fundamental problem in social ascertain source reliability and claim correctness without knowing either them priori, which referred as truth finding. Several key challenges exist order solve finding problem. First, neither nor collected data are known priori. Second, sources may make with different...

Many online social networking platforms are leveraging crowdsourcing to enhance the user experience. These seek incentivize heterogeneous workers exert efforts complete tasks (e.g., moderation of posts and articles) truthfully report their solutions. Output agreement mechanism majority voting) is a common approach this end. In an output mechanism, worker rewarded according whether his solution matches those peers. However, prior related work has not studied workers' accuracy how...

This paper presents a spatial-temporal aware analytical framework to solve the truth finding problem in social sensing applications. Social has emerged as new big data application paradigm of collecting observations about physical environment from sensors (e.g., humans) or devices on their behalf. The collected may be true false, and hence are viewed binary claims. A fundamental challenge applications lies accurately ascertaining correctness claims reliability sources without knowing either...

We study the design of incentive mechanisms for problem information elicitation without verification (IEWV). In IEWV, a data requester seeks to proper incentives optimize tradeoff between quality (collected from distributed crowd workers) and total cost (provided verifiable ground truth. While prior work often relies on sufficient knowledge worker information, we scenario where cannot access workers' heterogeneous costs ex-ante. propose continuum-armed bandit-based mechanism that dynamically...

Twitter has emerged as a new application paradigm of sensing the physical environment by using human sensors. These sensed observations are often viewed binary claims (either true or false). A fundamental challenge on is how to ascertain credibility and reliability sources without prior knowledge either them beforehand. This referred truth discovery. An important limitation exists in current Twitter-based discovery solutions: they did not explore theme relevance aspect correct identified...

This paper presents a spatial-temporal aware analytical framework to solve the truth finding problem in social sensing applications. Social has emerged as new big data application paradigm of collecting observations about physical environment from sensors (e.g., humans) or devices on their behalf. The collected may be true false, and hence are viewed binary claims. A fundamental challenge applications lies accurately ascertaining correctness claims reliability sources without knowing either...

In mobile crowdsourcing, platforms seek to incentivize heterogeneous workers complete tasks (e.g., road traffic sensing) and truthfully report their solutions. When cannot verify the quality of workers' solutions, crowdsourcing problem is known as <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">information elicitation without verification</i> (IEWV). an IEWV problem, a platform needs provide incentives motivate high-quality solutions truthful...

Crowdsourcing platforms often want to incentivize workers finish tasks with high quality and truthfully report their solutions. A solution requires a worker exert effort; platform can motivate such effort exertion truthful reporting by providing reward. We propose novel rewarding mechanism based on using truth detection technology, which verify the correctness of workers' responses questions an imperfect accuracy (e.g., regarding whether finishing they solutions). model interactions between...

We study a crowdsourcing problem where the platform aims to incentivize distributed workers provide high-quality and truthful solutions without ability verify solutions. While most prior work assumes that have symmetric information, we an asymmetric information scenario has informational advantages. Specifically, knows more regarding workers' average solution accuracy, can strategically reveal such workers. Workers will utilize announced determine likelihood they obtain reward if exerting...

We study a crowdsourcing problem, where platform aims to incentivize distributed workers provide high-quality and truthful solutions that are not verifiable. focus on largely overlooked yet pratically important asymmetric information scenario, the knows more regarding workers' average solution accuracy can strategically reveal such workers. Workers will utilize announced determine likelihood of obtaining reward. first case share same prior worker (but only observes realized value). consider...

Social sensing has emerged as a new data collection paradigm in networked applications where humans are used "sensors" to report their observations about the physical world. While many previous studies social focus on problem of ascertaining reliability sources and correctness reported claims (often known truth discovery), this paper investigates critical source selection. The goal is identify subset that can help effectively reduce computational complexity original discovery improve...

Cooperative content offloading is a promising technology to lessen heavy burden of wireless networks and improve the quality downloading services. Since few users are voluntary in providing free assistance, auction-based incentive mechanisms designed encourage participation. In existing mechanisms, each provider only acts as service seller. However, could also be partner requestor if having interest requested content. This dual identity can its cut down payment requestor. Based on this...

When it is difficult to verify contributed solutions in mobile crowdsourcing, the majority voting mechanism widely utilized incentivize distributed workers provide high-quality and truthful solutions. In mechanism, a worker rewarded based on whether his solution consistent with majority. However, most prior related work relies strong assumption that accuracy levels are public knowledge, which may not hold many practical scenarios. We relax such an propose online allows platform learn...

This paper presents a new principled framework for exploiting time-sensitive information to improve the truth discovery accuracy in social sensing applications. work is motivated by emergence of as paradigm collecting observations about physical environment from humans or devices on their behalf. These maybe true false, and hence are viewed binary claims. A fundamental problem applications lies ascertaining correctness claims reliability data sources. We refer this discovery. In paper, we...

With the burgeoning growth of online video platforms and escalating volume content, demand for proficient understanding tools has intensified markedly. Given remarkable capabilities Large Language Models (LLMs) in language multimodal tasks, this survey provides a detailed overview recent advancements harnessing power LLMs (Vid-LLMs). The emergent Vid-LLMs are surprisingly advanced, particularly their ability open-ended spatial-temporal reasoning combined with commonsense knowledge,...

This paper presents a scalable estimation theoretic framework to address the time-sensitive truth discovery problem with accuracy assessment in social sensing applications. Social has emerged as new application paradigm that provides us an unprecedented opportunity collect observations about physical world from humans or devices on their behalf. A fundamental challenge applications lies ascertaining correctness of claims and reliability data sources without knowing either them priori, which...

Social sensing has become a new crowdsourcing application paradigm where humans function as sensors to report their observations about the physical world. While many previous studies in social focus on problem of ascertaining reliability data sources and truthfulness reported claims (often known truth discovery), this paper investigates hypothesis validation goal is validate some high-level statements (referred hypotheses) from low-level claims) embedded data. The hypotheses cannot be...