Search

In this paper, we propose a two-stage ranking approach for recommending linear TV programs. The proposed approach first leverages user viewing patterns regarding time and TV channels to identify potential candidates for recommendation and then further leverages user preferences to rank these candidates given textual information about programs. To evaluate the method, we conduct empirical studies on a real-world TV dataset, the results of which demonstrate the superior performance of our model in terms of both recommendation accuracy and time efficiency., Comment: 8 pages

The Hitomi results for the Perseus cluster have shown that accurate atomic models are essential to the success of X-ray spectroscopic missions, and just as important as knowledge on instrumental calibration and astrophysical modeling. Preparing the models requires a multifaceted approach, including theoretical calculations, laboratory measurements, and calibration using real observations. In a previous paper, we presented a calculation of the electron impact cross sections on the transitions forming the Fe-L complex. In the present work, we systematically test the calculation against cross sections of ions measured in an electron beam ion trap experiment. A two-dimensional analysis in the electron beam energies and X-ray photon energies is utilized to disentangle radiative channels following dielectronic recombination, direct electron-impact excitation, and resonant excitation processes in the experimental data. The data calibrated through laboratory measurements are further fed into global modeling of the Chandra grating spectrum of Capella. We investigate and compare the fit quality, as well as sensitivity of the derived physical parameters to the underlying atomic data and the astrophysical plasma modeling. We further list the potential areas of disagreement between the observation and the present calculations, which in turn calls for renewed efforts in theoretical calculations and targeted laboratory measurements., Comment: Accepted for publication in A&A

We develop an automated technique to measure quasar redshifts in the Baryon Oscillation Spectroscopic Survey (BOSS) of the Sloan Digital Sky Survey (SDSS). Our technique is an extension of an earlier Gaussian process method for detecting damped Lyman-alpha absorbers (DLAs) in quasar spectra with known redshifts. We apply this technique to a subsample of SDSS DR12 with BAL quasars removed and redshift larger than 2.15. We show that we are broadly competitive to existing quasar redshift estimators, disagreeing with the PCA redshift by more than 0.5 in only 0.38% of spectra. Our method produces a probabilistic density function for the quasar redshift, allowing quasar redshift uncertainty to be propagated to downstream users. We apply this method to detecting DLAs, accounting in a Bayesian fashion for redshift uncertainty. Compared to our earlier method with a known quasar redshift, we have a moderate decrease in our ability to detect DLAs, predominantly in the noisiest spectra. The area under curve drops from 0.96 to 0.91. Our code is publicly available., Comment: 14 pages, 11 figures, accepted by MNRAS with some improvements and clarifications to discussion

In this paper, we propose a novel optimization criterion that leverages features of the skew normal distribution to better model the problem of personalized recommendation. Specifically, the developed criterion borrows the concept and the flexibility of the skew normal distribution, based on which three hyperparameters are attached to the optimization criterion. Furthermore, from a theoretical point of view, we not only establish the relation between the maximization of the proposed criterion and the shape parameter in the skew normal distribution, but also provide the analogies and asymptotic analysis of the proposed criterion to maximization of the area under the ROC curve. Experimental results conducted on a range of large-scale real-world datasets show that our model significantly outperforms the state of the art and yields consistently best performance on all tested datasets., Comment: Accepted by UAI'20. The first two authors contributed equally to this work; author order was determined by seniority

Conversational search plays a vital role in conversational information seeking. As queries in information seeking dialogues are ambiguous for traditional ad-hoc information retrieval (IR) systems due to the coreference and omission resolution problems inherent in natural language dialogue, resolving these ambiguities is crucial. In this paper, we tackle conversational passage retrieval (ConvPR), an important component of conversational search, by addressing query ambiguities with query reformulation integrated into a multi-stage ad-hoc IR system. Specifically, we propose two conversational query reformulation (CQR) methods: (1) term importance estimation and (2) neural query rewriting. For the former, we expand conversational queries using important terms extracted from the conversational context with frequency-based signals. For the latter, we reformulate conversational queries into natural, standalone, human-understandable queries with a pretrained sequence-tosequence model. Detailed analyses of the two CQR methods are provided quantitatively and qualitatively, explaining their advantages, disadvantages, and distinct behaviors. Moreover, to leverage the strengths of both CQR methods, we propose combining their output with reciprocal rank fusion, yielding state-of-the-art retrieval effectiveness, 30% improvement in terms of NDCG@3 compared to the best submission of TREC CAsT 2019., Comment: 28 pages. Accepted to ACM Transactions on Information Systems, Special Issue on Conversational Search and Recommendation. The first two authors contributed equally. Code: https://github.com/castorini/chatty-goose

This paper presents an empirical study of conversational question reformulation (CQR) with sequence-to-sequence architectures and pretrained language models (PLMs). We leverage PLMs to address the strong token-to-token independence assumption made in the common objective, maximum likelihood estimation, for the CQR task. In CQR benchmarks of task-oriented dialogue systems, we evaluate fine-tuned PLMs on the recently-introduced CANARD dataset as an in-domain task and validate the models using data from the TREC 2019 CAsT Track as an out-domain task. Examining a variety of architectures with different numbers of parameters, we demonstrate that the recent text-to-text transfer transformer (T5) achieves the best results both on CANARD and CAsT with fewer parameters, compared to similar transformer architectures.

We present a revised version of our automated technique using Gaussian processes (GPs) to detect Damped Lyman-$\alpha$ absorbers (DLAs) along quasar (QSO) sightlines. The main improvement is to allow our Gaussian process pipeline to detect multiple DLAs along a single sightline. Our DLA detections are regularised by an improved model for the absorption from the Lyman-$\alpha$ forest which improves performance at high redshift. We also introduce a model for unresolved sub-DLAs which reduces mis-classifications of absorbers without detectable damping wings. We compare our results to those of two different large-scale DLA catalogues and provide a catalogue of the processed results of our Gaussian process pipeline using 158 825 Lyman-$\alpha$ spectra from SDSS data release 12. We present updated estimates for the statistical properties of DLAs, including the column density distribution function (CDDF), line density ($dN/dX$), and neutral hydrogen density ($\Omega_{\textrm{DLA}}$)., Comment: 24 pages, 20 figures, plus 3 tables of figure values. Minor changes to match version published in MNRAS. Code available in https://github.com/rmgarnett/gp_dla_detection/tree/master/multi_dlas

We applied the T5 sequence-to-sequence model to tackle the AI2 WinoGrande Challenge by decomposing each example into two input text strings, each containing a hypothesis, and using the probabilities assigned to the "entailment" token as a score of the hypothesis. Our first (and only) submission to the official leaderboard yielded 0.7673 AUC on March 13, 2020, which is the best known result at this time and beats the previous state of the art by over five points.