We study a Dark Matter (DM) model in which the dominant coupling to the standard model occurs through a neutrino-DM-scalar coupling. The new singlet scalar will generically have couplings to nuclei/electrons arising from renormalizable Higgs portal interactions. As a result the DM particle $X$ can convert into a neutrino via scattering on a target nucleus $\mathcal{N}$: $ X + \mathcal{N} \rightarrow \nu + \mathcal{N}$, leading to striking signatures at direct detection experiments. Similarly, DM can be produced in neutrino scattering events at neutrino experiments: $ \nu + \mathcal{N} \rightarrow X + \mathcal{N}$, predicting spectral distortions at experiments such as COHERENT. Furthermore, the model allows for late kinetic decoupling of dark matter with implications for small-scale structure. At low masses, we find that COHERENT and late kinetic decoupling produce the strongest constraints on the model, while at high masses the leading constraints come from DM down-scattering at XENON1T and Borexino. Future improvement will come from CE$\nu$NS data, ultra-low threshold direct detection, and rare kaon decays.

The unprecedented volume and rate of transient events that will be discovered by the Large Synoptic Survey Telescope (LSST) demands that the astronomical community update its followup paradigm. Alert-brokers -- automated software system to sift through, characterize, annotate and prioritize events for followup -- will be critical tools for managing alert streams in the LSST era. The Arizona-NOAO Temporal Analysis and Response to Events System (ANTARES) is one such broker. In this work, we develop a machine learning pipeline to characterize and classify variable and transient sources only using the available multiband optical photometry. We describe three illustrative stages of the pipeline, serving the three goals of early, intermediate and retrospective classification of alerts. The first takes the form of variable vs transient categorization, the second, a multi-class typing of the combined variable and transient dataset, and the third, a purity-driven subtyping of a transient class. While several similar algorithms have proven themselves in simulations, we validate their performance on real observations for the first time. We quantitatively evaluate our pipeline on sparse, unevenly sampled, heteroskedastic data from various existing observational campaigns, and demonstrate very competitive classification performance. We describe our progress towards adapting the pipeline developed in this work into a real-time broker working on live alert streams from time-domain surveys.

The Arizona-NOAO Temporal Analysis and Response to Events System (ANTARES) is a joint effort of NOAO and the Department of Computer Science at the University of Arizona to build prototype software to process alerts from time-domain surveys, especially LSST, to identify those alerts that must be followed up immediately. Value is added by annotating incoming alerts with existing information from previous surveys and compilations across the electromagnetic spectrum and from the history of past alerts. Comparison against a knowledge repository of properties and features of known or predicted kinds of variable phenomena is used for categorization. The architecture and algorithms being employed are described.