搜索结果：Member

As generative models increasingly produce samples that are indistinguishable from human-created content, it becomes difficult to determine whether a given data point was part of a model's natural training set or was generated by the model itself, especially when models memorize and reproduce training data. We formalize this challenge as Member vs Generated Inference (MGI): given a sample and a target generative model, infer whether the sample is a true training member or a generated output of that model. Focusing on image generation, we show that existing membership inference methods systematically misclassify generated samples as training members, while attribution-based methods often misclassify true members as generated. This failure arises because both approaches rely on likelihood-related signals that are similarly elevated for training examples and for the model's own outputs. To address MGI, we propose Data Circuit Breaker (DCB), a three-stage method that combines complementary signals from a generative model's autoencoder and latent generator to distinguish training members from generated samples. Across multiple generative models, including image autoregressive and diffusi

As of 2022, the European Union has taken several steps regarding enlargement. We focus on the accession of countries with which the Union is actively negotiating membership. This is examined under two enlargement scenarios: first, the enlargement along the lines of the Western Balkan countries, and second, the accession of a trio (Ukraine, Moldova, and Georgia) to the already enlarged Union. We determine the a priori power of the member states based on the Banzhaf and Shapley--Shubik indices. Various coalitions are also assumed to assess the power and influence of member states, considering both pre- and post-enlargement scenarios. We found a rare case when the two indices give different rankings. In the case of the Western Balkan countries' accession, the smaller population member states gain power, presenting an example of the new member paradox. While in a Union of 36 members, every member state loses some of their current power. However, some coalitions are better off with the EU36 enlargement than a EU33 one.

In this paper, we study the characteristics of the member firms on the Korea Exchange. The member firms intermediate between the market participants and the exchange, and all the participants should trade stocks through members. To identify the characteristics of member firms, all member firms are categorized into three groups, such as the domestic members similar to individuals (DIMs), the domestic members similar to institutions (DSMs), and the foreign members (FRMs), in terms of the type of investor. We examine the dynamics of the member firms. The trading characteristics of members are revealed through the directionality and trend. While FRMs tend to trade one-way and move with the price change, DIMs are the opposite. In the market, DIMs and DSMs do herd and the herding moves in the opposite direction of the price change. One the other hand, FRMs do herd in the direction of the price change. The network analysis supports that the members are clustered into three groups similar to DIMs, DSMs, and FRMs. Finally, random matrix theory and a cross-sectional regression show that the inventory variation of members possesses significant information about stock prices and that member he

To identify member populations of IC 1396, we employ the random forest (RF) classifier of machine learning technique. Random forest classifier is an ensemble of individual decision trees suitable for large, high-dimensional datasets. The training set used in this work is derived from previous Gaia-based studies, where the member stars are younger than $\sim$ 10~Myr. However, its sensitivity is limited to $\sim$ 20~mag in the $\rm r_2$ band, making it challenging to identify candidates at the fainter end. In this analysis, in addition to magnitudes and colours, we incorporate several derived parameters from the magnitude and colour of the sources to identify candidate members of the star-forming complex. By employing this method, we are able to identify promising candidate member populations of the star-forming complex. We discuss the associated limitations and caveats in the method and for improvment in future studies. In this analysis, we identify 2425 high-probability low-mass stars distributed within the entire star-forming complex, of which 1331 are new detections. Comparison of these identified member populations shows a high retrieval rate with Gaia-based literature sources,

Many people browse online communities to learn from others' experiences and opinions, e.g., for constructing travel plans. Conversational search powered by large language models (LLMs) could ease this information-seeking task, but it remains under-investigated within the online community. In this paper, we first conducted an exploratory study (N=10) that indicated the helpfulness of a classic conversational search tool and identified room for improvement. Then, we proposed ConSearcher, an LLM-powered tool with dynamically generated member personas based on user queries to facilitate conversational search in the community. In ConSearcher, users can clarify their interests by checking what a simulated member similar to them may ask and get responses from diverse members' perspectives. A within-subjects study (N=27) showed that compared to two conversational search baselines, ConSearcher led to significantly higher information-seeking outcome and user engagement but raised concerns about over-personalization. We discuss implications for supporting conversational information seeking in online communities.

This paper develops a multi-objective optimization framework to analyze the trade-offs between annual costs and resilience in energy communities. Under this framework, three energy community operation strategies are analyzed: a reference case where all assets are member-owned, implementing a communal battery electric storage system, and subsidizing energy-poor members. The results indicate that increasing resilience leads to higher operational costs and smaller feasible ranges of energy community energy prices. The analysis reveals that those trade-offs have a heterogeneous impact across different member groups. Owners photovoltaics are most affected due to curtailed energy. Notably, the study shows that while implementing community-owned storage does not directly provide financial benefits to energy-poor members, alleviating the energy price for these members leads to an overall cost reduction of more than 30%. This research provides insights into the operational complexity of energy communities and highlights the importance of technologically robust and socially inclusive energy communities.

Many recent research studies have focused on the well-being of software development team members, as this aspect may be critical not only for productivity and performance at work but also for the physical health and personal life of employees. Many studies agree that an important factor of team member well-being is whether team members feel appreciated and acknowledged for their contributions. This paper presents the results of a project on the team well-being analysis as well as the prototype developed within the project.

A detailed study of stellar populations in Milky Way (MW) satellite galaxies remains an observational challenge due to their faintness and fewer spectroscopically confirmed member stars. We use unsupervised machine learning methods to identify new members for nine nearby MW satellite galaxies using Gaia data release-3 (Gaia DR3) astrometry and the Dark Energy Survey (DES) and the DECam Local Volume Exploration Survey (DELVE) photometry. Two density-based clustering algorithms, DBSCAN and HDBSCAN, have been used in the four-dimensional astrometric parameter space to identify member stars belonging to MW satellite galaxies. Our results indicate that we can recover more than 80% of the known spectroscopically confirmed members in most of the satellite galaxies and also reject 95-100% of spectroscopic non-members. We have also added many new members using this method. We compare our results with previous studies that also use photometric and astrometric data and discuss the suitability of density-based clustering methods for MW satellite galaxies

We present a large sample of new members of the Taurus star-forming region that extend from stellar to planetary masses. To identify candidate members at substellar masses, we have used color-magnitude diagrams and proper motions measured with several wide-field optical and infrared (IR) surveys. At stellar masses, we have considered the candidate members that were found in a recent analysis of high-precision astrometry from the Gaia mission. Using new and archival spectra, we have measured spectral types and assessed membership for these 161 candidates, 79 of which are classified as new members. Our updated census of Taurus now contains 519 known members. According to Gaia data, this census should be nearly complete for spectral types earlier than M6-M7 at A_J<1. For a large field encompassing ~72% of the known members, the census should be complete for K<15.7 at A_J<1.5, which corresponds to ~5-13 M_Jup for ages of 1-10 Myr based on theoretical evolutionary models. Our survey has doubled the number of known members at greater or equal to M9 and has uncovered the faintest known member in M_K, which should have a mass of ~3-10 M_Jup for ages of 1-10 Myr. We have used mid-I

The important task of developing verification system of data of virtual community member on the basis of computer-linguistic analysis of the content of a large sample of Ukrainian virtual communities is solved. The subject of research is methods and tools for verification of web-members socio-demographic characteristics based on computer-linguistic analysis of their communicative interaction results. The aim of paper is to verifying web-user personal data on the basis of computer-linguistic analysis of web-members information tracks. The structure of verification software for web-user profile is designed for a practical implementation of assigned tasks. The method of personal data verification of web-members by analyzing information track of virtual community member is conducted. For the first time the method for checking the authenticity of web members personal data, which helped to design of verification tool for socio-demographic characteristics of web-member is developed. The verification system of data of web-members, which forms the verified socio-demographic profiles of web-members, is developed as a result of conducted experiments. Also the user interface of the developed v

In this paper the results of matching the RAdial Velocity Experiment (RAVE), a spectroscopic Southern hemisphere survey (9 $ < I_{DENIS} <$ 12), and XMM-Newton Serendipitous Source Catalogue (3XMM) are presented. The latest data releases of RAVE and XMM were matched and a X-ray RAVE catalogue of 1071 stars was obtained. Then the catalogue was checked for possible Gould Belt (GB) members. We obtained a subsample of 10 stars that meet the GB membership criteria. This subsample and GB member candidates were tested photometrically and kinematically. Among the members there are two BY Dra type variables, an NGC2451 open cluster member, a high proper motion star. The rest are regular main sequence stars. The members have very low velocity dispersions which lead us to think that the members belong in a single structure. We also found out that a kinematical GB membership test might be possible to derive given a large enough GB member sample as they fit in a narrow interval in space velocity diagrams.

We investigate the possible presence of systematic rotation in the member galaxies of a sample of 17 nearby ($z<0.1$), rich (at least 80 identified members) Abell clusters. We also assess the extent to which low-number statistics may influence the recovery of the rotation parameters. Following the methods often used in the context of globular clusters and of clusters of galaxies, we estimate a representative value of the systematic rotation velocity and the position angle of the projected rotation axis for the set of spectroscopically confirmed member galaxies within 1.5 Mpc from the centre of each cluster. We study the robustness of our rotational velocity measurements as a function of the number of galaxies included in the analysis with a bootstrapping technique. Eight clusters with sufficiently abundant and regular data (A1367, A1650, A2029, A2065, A2142, A2199, A2255 and A2670) exhibit a significantly high rotational velocity, when compared to their velocity dispersion ($v_{rot}/σ\geq 0.15$). Interestingly, three of them (A1650, A2029 and A2199) are confirmed to be cool-core, relaxed clusters with no evidence of recent mergers, as suggested by X-ray observational data. We al

A 200-member ensemble developed at the Center for Western Weather and Water Extremes based on the Weather Research and Forecast atmospheric model tailored for the prediction of atmospheric rivers and associated heavy-to-extreme precipitation events over the Western US (West-WRF) is presented. The ensemble (WW200En) is generated with initial and boundary conditions from the US National Center for Environmental Prediction's Global Ensemble Forecast System (GEFS) and the European Centre for Medium-Range Weather Forecasts' Ensemble Prediction System (EPS), 100 unique combinations of microphysics, planetary boundary layer, and cumulus schemes, as well as perturbations applied to each of the 200 members based on the stochastic kinetic-energy backscatter scheme. Each member is run with 9-km horizontal increments and 60 vertical levels for a 10-month period spanning two winters. The performance of WW200En is compared to GEFS and EPS for probabilistic forecasts of 24-h precipitation, integrated water vapor transport (IVT), and for several thresholds including high percentiles of the observed climatological distribution. The WW200En precipitation forecast skill is better than GEFS at nearly

In this article, we study the inconsistency of a system of $\max$-product fuzzy relational equations and of a system of $\max$-Lukasiewicz fuzzy relational equations. For a system of $\max-\min$ fuzzy relational equations $A \Box_{\min}^{\max} x = b$ and using the $L_\infty$ norm, (Baaj, 2023) showed that the Chebyshev distance $Δ= \inf_{c \in \mathcal{C}} \Vert b - c \Vert$, where $\mathcal{C}$ is the set of second members of consistent systems defined with the same matrix $A$, can be computed by an explicit analytical formula according to the components of the matrix $A$ and its second member $b$. In this article, we give analytical formulas analogous to that of (Baaj, 2023) to compute the Chebyshev distance associated to the second member of a system of $\max$-product fuzzy relational equations and that associated to the second member of a system of $\max$-Lukasiewicz fuzzy relational equations.

Group decisions involve the combination of evidence accumulation by individual members and direct member-to-member interactions. We consider a simplified framework of two deciders, each undergoing a two alternative forced choice task, with the choices of early deciding members biasing members who have yet to choose. We model decision dynamics as a drift-diffusion process and present analysis of the associated Fokker-Planck equation for the group. We show that the probability of coordinated group decisions (both members make the same decision) is maximized by setting the decision threshold of one member to a lower value than its neighbor's. This result is akin to a speed-accuracy tradeoff, where the penalty of lowering the decision threshold is choice inaccuracy while the benefit is that earlier decisions have a higher probability of influencing the other member. We numerically extend these results to large group decisions, where it is shown that by choosing the appropriate parameters, a small but vocal component of the population can have a large amount of influence on the total system.

An academic team is a highly-cohesive collaboration group of scholars, which has been recognized as an effective way to improve scientific output in terms of both quality and quantity. However, the high staff turnover brings about a series of problems that may have negative influence on team performance. To address this challenge, we first detect the tendency of the member who may potentially leave. Here the outlierness is defined with respect to familiarity, which is quantified by using collaboration intensity. It is assumed that if a team member has a higher familiarity with scholars outside the team, then this member might probably leave the team. To minimize the influence caused by the leaving of such an outlier member, we propose an optimization solution to find a proper candidate who can replace the outlier member. Based on random walk with graph kernel, our solution involves familiarity matching, skill matching, as well as structure matching. The proposed approach proves to be effective and outperforms existing methods when applied to computer science academic teams.

We use new data from Gaia DR2 to confirm that the young L1 $γ$ candidate member of the TW Hya association (TWA) is now a bona fide member with a model-dependent mass estimate of ~15 $M_{Jup}$. The ambiguous M9 $γ$ candidate member 2MASS J12474428---3816464 also gets a higher Bayesian membership probability for TWA membership as a result of Gaia DR2 data and a new radial velocity measurement, but it remains unclear whether it is a true member of TWA or if it is an unrelated young interloper, because it is separated by 4.6 km/s from the locus of TWA members in UVW space.

We present optical photometry (i- and Z-band) and low-resolution spectroscopy (640-1015 nm) of very faint candidate members (J = 20.2-21.2 mag) of the Pleiades star cluster (120 Myr). The main goal is to address their cluster membership via photometric, astrometric, and spectroscopic studies, and to determine the properties of the least massive population of the cluster through the comparison of the data with younger and older spectral counterparts and state-of-the art model atmospheres. We confirm three bona-fide Pleiades members that have extremely red optical and infrared colors, effective temperatures of ~1150 K and ~1350 K, and masses in the interval 11-20 Mjup, and one additional likely member that shares the same motion as the cluster but does not appear to be as red as the other members with similar brightness. This latter object requires further near-infrared spectroscopy to fully address its membership in the Pleiades. The optical spectra of two bona-fide members were classified as L6-L7 and show features of KI, a tentative detection of CsI, hydrides and water vapor with an intensity similar to high-gravity dwarfs of related classification despite their young age. The pro

In membership inference attacks (MIAs), an adversary observes the predictions of a model to determine whether a sample is part of the model's training data. Existing MIA defenses conceal the presence of a target sample through strong regularization, knowledge distillation, confidence masking, or differential privacy. We propose MIAShield, a new MIA defense based on preemptive exclusion of member samples instead of masking the presence of a member. The key insight in MIAShield is weakening the strong membership signal that stems from the presence of a target sample by preemptively excluding it at prediction time without compromising model utility. To that end, we design and evaluate a suite of preemptive exclusion oracles leveraging model-confidence, exact or approximate sample signature, and learning-based exclusion of member data points. To be practical, MIAShield splits a training data into disjoint subsets and trains each subset to build an ensemble of models. The disjointedness of subsets ensures that a target sample belongs to only one subset, which isolates the sample to facilitate the preemptive exclusion goal. We evaluate MIAShield on three benchmark image classification da