Search for: All records

The capability to generate responses with diversity and faithfulness using factual knowledge is paramount for creating a human-like, trustworthy dialogue system. Common strategies either adopt a two-step paradigm, which optimizes knowledge selection and response generation separately and may overlook the inherent correlation between these two tasks, or leverage conditional variational method to jointly optimize knowledge selection and response generation by employing an inference network. In this paper, we present an end-to-end learning framework, termed Sequential Posterior Inference (SPI), capable of se- lecting knowledge and generating dialogues by approximately sampling from the posterior distribution. Unlike other methods, SPI does not require the inference network or assume a simple geometry of the posterior distribution. This straightforward and intuitive inference procedure of SPI directly queries the response generation model, allowing for accurate knowledge selection and generation of faithful responses. In addition to modeling contributions, our experimental results on two common dialogue datasets (Wizard of Wikipedia and Holl-E) demonstrate that SPI outperforms previous strong baselines according to both automatic and human evaluation metrics. 

We study the ionizing photon production efficiency at the end of the Epoch of Reionization (z ∼ 5.4 − 6.6) for a sample of 30 Ly α emitters. This is a crucial quantity to infer the ionizing photon budget of the universe. These objects were selected to have reliable spectroscopic redshifts, assigned based on the profile of their Ly α emission line, detected in the MUSE deep fields. We exploit medium-band observations from the JWST Extragalactic Medium-band Survey (JEMS) to find the flux excess corresponding to the redshifted Hα emission line. We estimate the ultraviolet (UV) luminosity by fitting the full JEMS photometry, along with several HST photometric points, with Prospector. We find a median UV continuum slope of $\beta = -2.09^{+0.23}_{-0.21}$, indicating young stellar populations with little-to-no dust attenuation. Supported by this, we derive ξion,0 with no dust attenuation and find a median value of log$\frac{\xi _{ion,0}}{\text{Hz erg}^{-1}} = 25.44^{+0.21}_{-0.15}$. If we perform dust attenuation corrections and assume a Calzetti attenuation law, our values are lowered by ∼0.1 dex. Our results suggest Ly α emitters at the Epoch of Reionization have slightly enhanced ξion,0 compared to previous estimations from literature, in particular, when compared to the non-Ly α emitting population. This initial study provides a promising outlook on the characterization of ionizing photon production in the early universe. In the future, a more extensive study will be performed on the entire data set provided by the JWST Advanced Deep Extragalactic Survey (JADES). Thus, for the first time, allowing us to place constraints on the wider galaxy populations driving reionization.

 

Abstract ATHENA has been designed as a general purpose detector capable of delivering the full scientific scope of the Electron-Ion Collider. Careful technology choices provide fine tracking and momentum resolution, high performance electromagnetic and hadronic calorimetry, hadron identification over a wide kinematic range, and near-complete hermeticity.This article describes the detector design and its expected performance in the most relevant physics channels. It includes an evaluation of detector technology choices, the technical challenges to realizing the detector and the R&D required to meet those challenges.