Search

Conversational information-seeking (CIS) is an emerging paradigm for knowledge acquisition and exploratory search. Traditional web search interfaces enable easy exploration of entities, but this is limited in conversational settings due to the limited-bandwidth interface. This paper explore ways to rewrite answers in CIS, so that users can understand them without having to resort to external services or sources. Specifically, we focus on salient entities -- entities that are central to understanding the answer. As our first contribution, we create a dataset of conversations annotated with entities for saliency. Our analysis of the collected data reveals that the majority of answers contain salient entities. As our second contribution, we propose two answer rewriting strategies aimed at improving the overall user experience in CIS. One approach expands answers with inline definitions of salient entities, making the answer self-contained. The other approach complements answers with follow-up questions, offering users the possibility to learn more about specific entities. Results of a crowdsourcing-based study indicate that rewritten answers are clearly preferred over the original ones. We also find that inline definitions tend to be favored over follow-up questions, but this choice is highly subjective, thereby providing a promising future direction for personalization.

While the body of research directed towards constructing and generating clarifying questions in mixed-initiative conversational search systems is vast, research aimed at processing and comprehending users' answers to such questions is scarce. To this end, we present a simple yet effective method for processing answers to clarifying questions, moving away from previous work that simply appends answers to the original query and thus potentially degrades retrieval performance. Specifically, we propose a classifier for assessing usefulness of the prompted clarifying question and an answer given by the user. Useful questions or answers are further appended to the conversation history and passed to a transformer-based query rewriting module. Results demonstrate significant improvements over strong non-mixed-initiative baselines. Furthermore, the proposed approach mitigates the performance drops when non useful questions and answers are utilized., Comment: This is the author's version of the work. The definitive version is published in: Proceedings of the 46th European Conference on Information Retrieval (ECIR '24), March 24-28, 2024, Glasgow, Scotland

Clarifying user's information needs is an essential component of modern search systems. While most of the approaches for constructing clarifying prompts rely on query facets, the impact of the quality of the facets is relatively unexplored. In this work, we concentrate on facet quality through the notion of facet coherency and assess its importance for overall usefulness for clarification in search. We find that existing evaluation procedures do not account for facet coherency, as evident by the poor correlation of coherency with automated metrics. Moreover, we propose a coherency classifier and assess the prevalence of incoherent facets in a well-established dataset on clarification. Our findings can serve as motivation for future work on the topic.

Recent studies show that Generative Relevance Feedback (GRF), using text generated by Large Language Models (LLMs), can enhance the effectiveness of query expansion. However, LLMs can generate irrelevant information that harms retrieval effectiveness. To address this, we propose Generative Relevance Modeling (GRM) that uses Relevance-Aware Sample Estimation (RASE) for more accurate weighting of expansion terms. Specifically, we identify similar real documents for each generated document and use a neural re-ranker to estimate their relevance. Experiments on three standard document ranking benchmarks show that GRM improves MAP by 6-9% and R@1k by 2-4%, surpassing previous methods.

This research aims to explore various methods for assessing user feedback in mixed-initiative conversational search (CS) systems. While CS systems enjoy profuse advancements across multiple aspects, recent research fails to successfully incorporate feedback from the users. One of the main reasons for that is the lack of system-user conversational interaction data. To this end, we propose a user simulator-based framework for multi-turn interactions with a variety of mixed-initiative CS systems. Specifically, we develop a user simulator, dubbed ConvSim, that, once initialized with an information need description, is capable of providing feedback to a system's responses, as well as answering potential clarifying questions. Our experiments on a wide variety of state-of-the-art passage retrieval and neural re-ranking models show that effective utilization of user feedback can lead to 16% retrieval performance increase in terms of nDCG@3. Moreover, we observe consistent improvements as the number of feedback rounds increases (35% relative improvement in terms of nDCG@3 after three rounds). This points to a research gap in the development of specific feedback processing modules and opens a potential for significant advancements in CS. To support further research in the topic, we release over 30,000 transcripts of system-simulator interactions based on well-established CS datasets., Comment: 11 pages, 2 figures, to be published in SIGIR 2023

We report the discovery of high velocity candidates among RR~Lyrae stars found in the Milky Way halo. We identified 9 RR~Lyrae stars with Galactocentric velocities exceeding the local escape velocity based on the assumed Galaxy potential. Based on close examination of their orbits', we ruled out their ejection location in the Milky Way disk and bulge. The spatial distribution revealed that seven out of 9 pulsators overlap with the position of the Sagittarius stellar stream. Two out of these seven RR~Lyrae stars can be tentatively linked to the Sagittarius dwarf spheroidal galaxy on the basis of their orbits. Focusing on the high-velocity tail of the RR~Lyrae velocity distribution we estimate the escape velocity in the Solar neighborhood to be $v_{\rm esc}=512^{+94}_{-37}$\,km\,s$^{-1}$~($4$ to $12$\,kpc), and beyond the Solar neighborhood as $v_{\rm esc}=436^{+44}_{-22}$\,km\,s$^{-1}$~and $v_{\rm esc}=393^{+53}_{-26}$\,km\,s$^{-1}$~(for distances between $12$ to $20$\,kpc and $20$ to $28$\,kpc), respectively. We utilized three escape velocity estimates together with the local circular velocity to estimate the Milky Way mass. The resulting measurement $M_{\rm 200}=0.83^{+0.29}_{-0.16} \cdot 10^{12}$\,M$_{\odot}$ falls on the lower end of the current Milky Way mass estimates, but once corrected for the likely bias in the escape velocity (approximately $10$ percent increase of the escape velocity), our mass estimate yields $M_{\rm 200}=1.26^{+0.40}_{-0.22} \cdot 10^{12}$\,M$_{\odot}$, which is in agreement with estimates based on different diagnostics of the Milky Way mass. The MW mass within $20$\,kpc then corresponds to $M_{\rm MW} \left(r < 20\,\text{kpc} \right)=1.9^{+0.2}_{-0.1} \times 10^{11}$\,M$_{\odot}$ without correction for bias, and $M_{\rm MW} \left(r < 20\,\text{kpc} \right)=2.1^{+0.2}_{-0.1} \times 10^{11}$\,M$_{\odot}$ corrected for a likely offset in escape velocities., Comment: Accepted for publication in A&A

Clarifying the underlying user information need by asking clarifying questions is an important feature of modern conversational search system. However, evaluation of such systems through answering prompted clarifying questions requires significant human effort, which can be time-consuming and expensive. In this paper, we propose a conversational User Simulator, called USi, for automatic evaluation of such conversational search systems. Given a description of an information need, USi is capable of automatically answering clarifying questions about the topic throughout the search session. Through a set of experiments, including automated natural language generation metrics and crowdsourcing studies, we show that responses generated by USi are both inline with the underlying information need and comparable to human-generated answers. Moreover, we make the first steps towards multi-turn interactions, where conversational search systems asks multiple questions to the (simulated) user with a goal of clarifying the user need. To this end, we expand on currently available datasets for studying clarifying questions, i.e., Qulac and ClariQ, by performing a crowdsourcing-based multi-turn data acquisition. We show that our generative, GPT2-based model, is capable of providing accurate and natural answers to unseen clarifying questions in the single-turn setting and discuss capabilities of our model in the multi-turn setting. We provide the code, data, and the pre-trained model to be used for further research on the topic.

We present new period-$\phi_{31}$-[Fe/H] relations for first overtone RRL stars (RRc), calibrated over a broad range of metallicities ($-2.5 < \textrm{[Fe/H]}< 0.0$) utilizing the largest currently available set of Galactic halo field RRL with homogeneous spectroscopic metallicities. Our relations are defined in the optical (ASAS-SN $V$-band) and, inaugurally, in the infrared (WISE $W1$ and $W2$ bands). Our $V$-band relation can reproduce individual RRc spectroscopic metallicities with a dispersion of 0.30 dex over the entire metallicity range of our calibrator sample (an RMS smaller than what we found for other relations in literature including non-linear terms). Our infrared relation has a similar dispersion in the low and intermediate metallicity range ($\textrm{[Fe/H]} < -0.5$) but tends to underestimate the [Fe/H] abundance around solar metallicity. We tested our relations by measuring both the metallicity of the Sculptor dSph and a sample of Galactic globular clusters, rich in both RRc and RRab stars. The average metallicity we obtain for the combined RRL sample in each cluster is within $\pm 0.08$ dex of their spectroscopic metallicities. The infrared and optical relations presented in this work will enable deriving reliable photometric RRL metallicities in conditions where spectroscopic measurements are not feasible; e.g., in distant galaxies or reddened regions (observed with upcoming Extremely Large Telescopes and the James Webb Space Telescope), or in the large sample of new RRL that will be discovered in large-area time-domain photometric surveys (such as LSST and the Roman space telescope)., Comment: Accepted by ApJ, 22 pages, 9 Figures, 3 Tables

Classical Cepheids are the most popular distance indicators and tracers of young stellar populations. The key advantage is that they are bright and they can be easily identified in Local Group and Local Volume galaxies. Their evolutionary and pulsation properties depend on their chemical abundances. The main aim of this investigation is to perform a new and accurate abundance analysis of two tens of calibrating Galactic Cepheids using high spectral resolution (R$\sim$40,000-115,000) and high S/N spectra ($\sim$400) covering the entire pulsation cycle. We focus our attention on possible systematics affecting the estimate of atmospheric parameters and elemental abundances along the pulsation cycle. We cleaned the line list by using atomic transition parameters based on laboratory measurements and by removing lines that are either blended or display abundance variations along the pulsation cycle. The spectroscopic approach that we developed brings forward small dispersions in the variation of the atmospheric parameters ($\sigma$($T_{\rm eff}$)$\sim$50 K, $\sigma$($\log{g}$)$\sim$0.2 dex, and $\sigma$($\xi$)$\sim$0.2 km/s) and in the abundance of both iron ($\lesssim$ 0.05 dex) and alpha elements ($\lesssim$0.10 dex) over the entire pulsation cycle. We also provide new and accurate effective temperature templates by splitting the calibrating Cepheids into four different period bins, ranging from short to long periods. For each period bin, we performed an analytical fit with Fourier series providing $\theta = 5040/{T_{\rm eff}}$ as a function of the pulsation phase. The current findings are a good viaticum to trace the chemical enrichment of the Galactic thin disk by using classical Cepheids and a fundamental stepping stone for further investigations into the more metal-poor regime typical of Magellanic Cepheids., Comment: 31 pages, 23 figures (out of which 11 are in the appendices A and B), to be published in the Astronomy & Astrophysics journal

Due to the worldwide accessibility to the Internet along with the continuous advances in mobile technologies, physical and digital worlds have become completely blended, and the proliferation of social media platforms has taken a leading role over this evolution. In this paper, we undertake a thorough analysis towards better visualising and understanding the factors that characterise and differentiate social media users affected by mental disorders. We perform different experiments studying multiple dimensions of language, including vocabulary uniqueness, word usage, linguistic style, psychometric attributes, emotions' co-occurrence patterns, and online behavioural traits, including social engagement and posting trends. Our findings reveal significant differences on the use of function words, such as adverbs and verb tense, and topic-specific vocabulary, such as biological processes. As for emotional expression, we observe that affected users tend to share emotions more regularly than control individuals on average. Overall, the monthly posting variance of the affected groups is higher than the control groups. Moreover, we found evidence suggesting that language use on micro-blogging platforms is less distinguishable for users who have a mental disorder than other less restrictive platforms. In particular, we observe on Twitter less quantifiable differences between affected and control groups compared to Reddit., Comment: To appear in Elsevier Information Processing & Management

As the popularity of Location-based Social Networks (LBSNs) increases, designing accurate models for Point-of-Interest (POI) recommendation receives more attention. POI recommendation is often performed by incorporating contextual information into previously designed recommendation algorithms. Some of the major contextual information that has been considered in POI recommendation are the location attributes (i.e., exact coordinates of a location, category, and check-in time), the user attributes (i.e., comments, reviews, tips, and check-in made to the locations), and other information, such as the distance of the POI from user's main activity location, and the social tie between users. The right selection of such factors can significantly impact the performance of the POI recommendation. However, previous research does not consider the impact of the combination of these different factors. In this paper, we propose different contextual models and analyze the fusion of different major contextual information in POI recommendation. The major contributions of this paper are: (i) providing an extensive survey of context-aware location recommendation (ii) quantifying and analyzing the impact of different contextual information (e.g., social, temporal, spatial, and categorical) in the POI recommendation on available baselines and two new linear and non-linear models, that can incorporate all the major contextual information into a single recommendation model, and (iii) evaluating the considered models using two well-known real-world datasets. Our results indicate that while modeling geographical and temporal influences can improve recommendation quality, fusing all other contextual information into a recommendation model is not always the best strategy., Comment: To appear in ACM TOIS

Metal-deficient stars are important tracers for understanding the early formation of the Galaxy. Recent large-scale surveys with both photometric and spectroscopic data have reported an increasing number of metal-deficient stars whose kinematic features are consistent with those of the disk stellar populations. We report the discovery of an RR~Lyrae variable (hereafter RRL) that is located within the thick disk and has an orbit consistent with the thick-disk kinematics. Our target RRL (HD 331986) is located at around 1 kpc from the Sun and, with V=11.3, is among the 130 brightest RRLs known so far. However, this object was scarcely studied because it is in the midplane of the Galaxy, the Galactic latitude around -1 deg. Its near-infrared spectrum (0.91-1.32 micron) shows no absorption line except hydrogen lines of the Paschen series, suggesting [Fe/H] less than -2.5. It is the most metal-deficient RRL, at least, among the RRLs whose orbits are consistent with the disk kinematics, although we cannot determine to which of the disk and the halo it belongs. This unique RRL would provide us with essential clues for studying the early formation of stars in the inner Galaxy with further investigations, including high-resolution optical spectroscopy., Comment: 13 pages, 7 figures, 6 tables, accepted for publication in ApJ

This paper proposes an approach to aggregate nonstructured power system controllers preserving the dynamical characteristics of the original devices. The method is based on linear operations that use the frequency response of the elements, resulting in an accurate input-output description of the equivalent controller when compared to the original ones. The developed method was applied to a model of the future interconnected Paraguayan-Argentinean power system to produce a dynamic equivalent used in a real-time simulator to test the special protection scheme needed for the safe operation of the this future system. Transient and small-signal stability studies presented matching simulation results in the time domain with significantly reduced computational burden and processing time., Comment: Preprint submitted to the 22nd Power Systems Computation Conference (PSCC 2022)

Recent developments in the mobile app industry have resulted in various types of mobile apps, each targeting a different need and a specific audience. Consequently, users access distinct apps to complete their information need tasks. This leads to the use of various apps not only separately, but also collaboratively in the same session to achieve a single goal. Recent work has argued the need for a unified mobile search system that would act as metasearch on users' mobile devices. The system would identify the target apps for the user's query, submit the query to the apps, and present the results to the user in a unified way. In this work, we aim to deepen our understanding of user behavior while accessing information on their mobile phones by conducting an extensive analysis of various aspects related to the search process. In particular, we study the effect of task type and user demographics on their behavior in interacting with mobile apps. Our findings reveal trends and patterns that can inform the design of a more effective mobile information access environment., Comment: FQAS Invited Paper

Recent research has shown that mixed-initiative conversational search, based on the interaction between users and computers to clarify and improve a query, provides enormous advantages. Nonetheless, incorporating additional information provided by the user from the conversation poses some challenges. In fact, further interactions could confuse the system as a user might use words irrelevant to the information need but crucial for correct sentence construction in the context of multi-turn conversations. To this aim, in this paper, we have collected two conversational keyword extraction datasets and propose an end-to-end document retrieval pipeline incorporating them. Furthermore, we study the performance of two neural keyword extraction models, namely, BERT and sequence to sequence, in terms of extraction accuracy and human annotation. Finally, we study the effect of keyword extraction on the end-to-end neural IR performance and show that our approach beats state-of-the-art IR models. We make the two datasets publicly available to foster research in this area., Comment: Accepted in IIR 2021

We discuss the largest and most homogeneous spectroscopic dataset of field RR Lyrae variables (RRLs) available to date. We estimated abundances using both high-resolution and low-resolution ({\Delta S} method) spectra for fundamental (RRab) and first overtone (RRc) RRLs. The iron abundances for 7,941 RRLs were supplemented with similar literature estimates available, ending up with 9,015 RRLs (6,150 RRab, 2,865 RRc). The metallicity distribution shows a mean value of <[Fe/H]> = -1.51\pm0.01, and {\sigma}(standard deviation)= 0.41 dex with a long metal-poor tail approaching [Fe/H] = -3 and a sharp metal-rich tail approaching solar iron abundance. The RRab variables are more metal-rich (<[Fe/H]>ab = -1.48\pm0.01, {\sigma} = 0.41 dex) than RRc variables (<[Fe/H]>c = -1.58\pm0.01, {\sigma} = 0.40 dex). The relative fraction of RRab variables in the Bailey diagram (visual amplitude vs period) located along the short-period (more metal-rich) and the long-period (more metal-poor) sequences are 80% and 20\%, while RRc variables display an opposite trend, namely 30\% and 70\%. We found that the pulsation period of both RRab and RRc variables steadily decreases when moving from the metal-poor to the metal-rich regime. The visual amplitude shows the same trend, but RRc amplitudes are almost two times more sensitive than RRab amplitudes to metallicity. We also investigated the dependence of the population ratio (Nc/Ntot) of field RRLs on the metallicity and we found that the distribution is more complex than in globular clusters. The population ratio steadily increases from ~0.25 to ~0.36 in the metal-poor regime, it decreases from ~0.36 to ~0.18 for -1.8 < [Fe/H] < -0.9 and it increases to a value of ~0.3 approaching solar iron abundance., Comment: 22 pages, 13 figures, 3 tables. Accepted for publication in ApJ

We collected the largest spectroscopic catalog of RR Lyrae (RRLs) including $\approx$20,000 high-, medium- and low-resolution spectra for $\approx$10,000 RRLs. We provide the analytical forms of radial velocity curve (RVC) templates. These were built using 36 RRLs (31 fundamental -- split into three period bins -- and 5 first overtone pulsators) with well-sampled RVCs based on three groups of metallic lines (Fe, Mg, Na) and four Balmer lines (H$_\alpha$, H$_\beta$, H$_\gamma$, H$_\delta$). We tackled the long-standing problem of the reference epoch to anchor light curve and RVC templates. For the $V$-band, we found that the residuals of the templates anchored to the phase of the mean magnitude along the rising branch are $\sim$35\% to $\sim$45\% smaller than those anchored to the phase of maximum light. For the RVC, we used two independent reference epochs for metallic and Balmer lines and we verified that the residuals of the RVC templates anchored to the phase of mean RV are from 30\% (metallic lines) up to 45\% (Balmer lines) smaller than those anchored to the phase of minimum RV. We validated our RVC templates by using both the single- and the three-phase points approach. We found that barycentric velocities based on our RVC templates are two-three times more accurate than those available in the literature. We applied the current RVC templates to Balmer lines RVs of RRLs in the globular NGC~3201 collected with MUSE at VLT. We found the cluster barycentric RV of $V_{\gamma}$=496.89$\pm$8.37(error)$\pm$3.43 (standard deviation) km/s, which agrees well with literature estimates.

We provide the largest and most homogeneous sample of $\alpha$-element (Mg, Ca, Ti) and iron abundances for field RR Lyrae (RRLs, 162 variables) by using high-resolution spectra. The current measurements were complemented with similar abundances available in the literature for 46 field RRLs brought to our metallicity scale. We ended up with a sample of old (t$\ge$ 10 Gyr), low-mass stellar tracers (208 RRLs: 169 fundamental, 38 first overtone, 1 mixed mode) covering three dex in iron abundance (-3.00$\le$[Fe/H]$\le$0.24). We found that field RRLs are $\sim$0.3 dex more $\alpha$-poor than typical Halo tracers in the metal-rich regime, ([Fe/H]$\ge$-1.2) while in the metal-poor regime ([Fe/H]$\le$-2.2) they seem to be on average $\sim$0.1 dex more $\alpha$-enhanced. This is the first time that the depletion in $\alpha$-elements for solar iron abundances is detected on the basis of a large, homogeneous and coeval sample of old stellar tracers. Interestingly, we also detected a close similarity in the [$\alpha$/Fe] trend between $\alpha$-poor, metal-rich RRLs and red giants (RGs) in the Sagittarius dwarf galaxy as well as between $\alpha$-enhanced, metal-poor RRLs and RGs in ultra faint dwarf galaxies. These results are supported by similar elemental abundances for 46 field Horizontal Branch (HB) stars. These stars share with RRLs the same evolutionary phase and the same progenitors. This evidence further supports the key role that old stellar tracers play in constraining the early chemical enrichment of the Halo and, in particular, in investigating the impact that dwarf galaxies have had in the mass assembly of the Galaxy.

Accurate metallicities of RR Lyrae are extremely important in constraining period-luminosity-metallicity relationships (PLZ), particularly in the near-infrared. We analyse 69 high-resolution spectra of Galactic RR Lyrae stars from the Southern African Large Telescope (SALT). We measure metallicities of 58 of these RR Lyrae stars with typical uncertainties of 0.13 dex. All but one RR Lyrae in this sample has accurate ({\sigma}_parallax ~ 10%) parallax from Gaia. Combining these new high resolution spectroscopic abundances with similar determinations from the literature for 93 stars, we present new PLZ relationships in WISE W1 and W2 magnitudes, and the Wesenheit magnitudes W(W1,V-W1) and W(W2,V-W2)., Comment: 16 pages, 13 figures, accepted in MNRAS

We present newly-calibrated period-$\phi_{31}$-[Fe/H] relations for fundamental mode RR Lyrae stars in the optical and, for the first time, mid-infrared. This work's calibration dataset provides the largest and most comprehensive span of parameter space to date with homogeneous metallicities from $-3<\textrm{[Fe/H]}<0.4$ and accurate Fourier parameters derived from 1980 ASAS-SN ($V$-band) and 1083 WISE (NEOWISE extension, $W1$ and $W2$ bands) RR Lyrae stars with well-sampled light curves. We compare our optical period-$\phi_{31}$-[Fe/H] with those available in the literature and demonstrate that our relation minimizes systematic trends in the lower and higher metallicity range. Moreover, a direct comparison shows that our optical photometric metallicities are consistent with both those from high-resolution spectroscopy and globular clusters, supporting the good performance of our relation. We found an intrinsic scatter in the photometric metallicities (0.41 dex in the $V$-band and 0.50 dex in the infrared) by utilizing large calibration datasets covering a broad metallicity range. This scatter becomes smaller when optical and infrared bands are used together (0.37 dex). Overall, the relations derived in this work have many potential applications, including large-area photometric surveys with JWST in the infrared and LSST in the optical., Comment: Accepted by ApJ, 29 pages, 14 Figures, 4 Tables

Clarification is increasingly becoming a vital factor in various topics of information retrieval, such as conversational search and modern Web search engines. Prompting the user for clarification in a search session can be very beneficial to the system as the user's explicit feedback helps the system improve retrieval massively. However, it comes with a very high risk of frustrating the user in case the system fails in asking decent clarifying questions. Therefore, it is of great importance to determine when and how to ask for clarification. To this aim, in this work, we model search clarification prediction as user engagement problem. We assume that the better a clarification is, the higher user engagement with it would be. We propose a Transformer-based model to tackle the task. The comparison with competitive baselines on large-scale real-life clarification engagement data proves the effectiveness of our model. Also, we analyse the effect of all result page elements on the performance and find that, among others, the ranked list of the search engine leads to considerable improvements. Our extensive analysis of task-specific features guides future research.

We present a chemo-dynamical study of the Orphan stellar stream using a catalog of RR~Lyrae pulsating variable stars for which photometric, astrometric, and spectroscopic data are available. Employing low-resolution spectra from the Sloan Digital Sky Survey (SDSS), we determined line-of-sight velocities for individual exposures and derived the systemic velocities of the RR~Lyrae stars. In combination with the stars' spectroscopic metallicities and \textit{Gaia} EDR3 astrometry, we investigated the northern part of the Orphan stream. In our probabilistic approach, we found 20 single mode RR~Lyrae variables likely associated with the Orphan stream based on their positions, proper motions, and distances. The acquired sample permitted us to expand our search to nonvariable stars in the SDSS dataset, utilizing line-of-sight velocities determined by the SDSS. We found 54 additional nonvariable stars linked to the Orphan stream. The metallicity distribution for the identified red giant branch stars and blue horizontal branch stars is, on average, $-2.13\pm0.05$ dex and $-1.87\pm0.14$ dex, with dispersions of 0.23 and 0.43dex, respectively. The metallicity distribution of the RR~Lyrae variables peaks at $-1.80\pm0.06$ dex and a dispersion of 0.25dex. Using the collected stellar sample, we investigated a possible link between the ultra-faint dwarf galaxy Grus II and the Orphan stream. Based on their kinematics, we found that both the stream RR~Lyrae and Grus II are on a prograde orbit with similar orbital properties, although the large uncertainties on the dynamical properties render an unambiguous claim of connection difficult. At the same time, the chemical analysis strongly weakens the connection between both. We argue that Grus II in combination with the Orphan stream would have to exhibit a strong inverse metallicity gradient, which to date has not been detected in any Local Group system., Comment: 22 pages, 12 figures, Accepted for publication in A&A, final version

Users install many apps on their smartphones, raising issues related to information overload for users and resource management for devices. Moreover, the recent increase in the use of personal assistants has made mobile devices even more pervasive in users' lives. This paper addresses two research problems that are vital for developing effective personal mobile assistants: target apps selection and recommendation. The former is the key component of a unified mobile search system: a system that addresses the users' information needs for all the apps installed on their devices with a unified mode of access. The latter, instead, predicts the next apps that the users would want to launch. Here we focus on context-aware models to leverage the rich contextual information available to mobile devices. We design an in situ study to collect thousands of mobile queries enriched with mobile sensor data (now publicly available for research purposes). With the aid of this dataset, we study the user behavior in the context of these tasks and propose a family of context-aware neural models that take into account the sequential, temporal, and personal behavior of users. We study several state-of-the-art models and show that the proposed models significantly outperform the baselines., Comment: Accepted to ACM TOIS, 30 pages