Search

The handling of communication breakdowns and loss of engagement is an important aspect of spoken dialogue systems, particularly for chatting systems such as attentive listening, where the user is mostly speaking. We presume that a human is best equipped to handle this task and rescue the flow of conversation. To this end, we propose a semi-autonomous system, where a remote operator can take control of an autonomous attentive listening system in real-time. In order to make human intervention easy and consistent, we introduce automatic detection of low interest and engagement to provide explicit takeover prompts to the remote operator. We implement this semi-autonomous system which detects takeover points for the operator and compare it to fully tele-operated and fully autonomous attentive listening systems. We find that the semi-autonomous system is generally perceived more positively than the autonomous system. The results suggest that identifying points of conversation when the user starts to lose interest may help us improve a fully autonomous dialogue system.

The Extreme Universe Space Observatory on a Super Pressure Balloon 1 (EUSO-SPB1) was launched in 2017 April from Wanaka, New Zealand. The plan of this mission of opportunity on a NASA super pressure balloon test flight was to circle the southern hemisphere. The primary scientific goal was to make the first observations of ultra-high-energy cosmic-ray extensive air showers (EASs) by looking down on the atmosphere with an ultraviolet (UV) fluorescence telescope from suborbital altitude (33~km). After 12~days and 4~hours aloft, the flight was terminated prematurely in the Pacific Ocean. Before the flight, the instrument was tested extensively in the West Desert of Utah, USA, with UV point sources and lasers. The test results indicated that the instrument had sensitivity to EASs of approximately 3 EeV. Simulations of the telescope system, telescope on time, and realized flight trajectory predicted an observation of about 1 event assuming clear sky conditions. The effects of high clouds were estimated to reduce this value by approximately a factor of 2. A manual search and a machine-learning-based search did not find any EAS signals in these data. Here we review the EUSO-SPB1 instrument and flight and the EAS search., Comment: 18 pages, 19 figures

We study the implications of the stringy space-time uncertainty relation (STUR) for inflationary cosmology. By demanding that no fluctuation modes that exit the Hubble radius are affected by the nonlocality resulting from the STUR, we find an upper bound on the number of e-foldings of inflation. The bound is a factor of 2 weaker than what results from the Trans-Planckian Censorship Criterion (TCC). By demanding that the inflationary phase is simultaneously consistent with STUR and sufficiently long for inflation to provide a causal explanation of structure on the scale of the current Hubble radius, we find an upper bound on the energy scale of inflation. The bound is less restrictive than what follows from the TCC, but it remains in conflict with canonical single-field inflation models., Comment: 7 pages, 1 figure

Sila-cyclic rings are a class of organosilicon cyclic compounds and have abundant application in organic chemistry and materials science. However, it is still challenging to synthesize compounds with sila-cyclic rings in solution chemistry due to their low solubility and high reactivity. Recently, on-surface synthesis was introduced into organosilicon chemistry as 1,4- disilabenzene bridged nanostructures were obtained via coupling between bromo-substituted molecules and silicon atoms on Au(111). Here, we extend this strategy for syntheses of silole derivatives and graphene nanoribbons with eight-membered sila-cyclic rings from 2,2',6,6'- tetrabromobiphenyl and 1,4,5,8-tetrabromonaphthalene on Au(111), respectively. Their structures and electronic properties were investigated by a combination of scanning tunneling microscopy/spectroscopy and density functional theory calculations. This work demonstrates a generality of this synthesis strategy to fabricate various silicon incorporated nanostructures.

Dozens of planets are now discovered with large orbital obliquity, and have become the proof for the dynamical evolution of planetary orbits. In the current samples, there is an apparent clustering of planets around $90^\circ$, and also an absence of planets around $180^\circ$ although the latter is expected by some theories. Statistical extrapolation using Hierarchical Bayesian Analysis have recently refuted the significant clustering around $90^\circ$ and suggested that the distribution may actually be broader. In this work, the symmetric TESS transit light curve of KELT-19Ab is analyzed using gravity darkening to measure its true obliquity. Its large sky projected obliquity $\lambda = -179.7^{\circ+3.7^\circ}_{\,\,-3.8^\circ}$ makes KELT-19Ab the only currently known planet with obliquity potentially close to $180^\circ$. We apply spectroscopic constraints on $v\mathrm{sin}i$ and $\lambda$ as well as theoretical constraints on the limb-darkening coefficients to find that the KELT-19Ab's obliquity is $\psi = 155^{\circ+17^\circ}_{\,\,-21^\circ}$, in favor of a flipped orbit. The result is consistent with the statistically inferred uniformity of obliquity distribution, and also highlights the applicability of the gravity darkening technique to symmetric light curves., Comment: 12 pages, 6 figures. Accepted for publication in MNRAS

Effective extraction and application of linguistic features are central to the enhancement of spoken Language IDentification (LID) performance. With the success of recent large models, such as GPT and Whisper, the potential to leverage such pre-trained models for extracting linguistic features for LID tasks has become a promising area of research. In this paper, we explore the utilization of the decoder-based network from the Whisper model to extract linguistic features through its generative mechanism for improving the classification accuracy in LID tasks. We devised two strategies - one based on the language embedding method and the other focusing on direct optimization of LID outputs while simultaneously enhancing the speech recognition tasks. We conducted experiments on the large-scale multilingual datasets MLS, VoxLingua107, and CommonVoice to test our approach. The experimental results demonstrated the effectiveness of the proposed method on both in-domain and out-of-domain datasets for LID tasks., Comment: Accepted by IEEE ASRU2023

Multi-talker overlapped speech recognition remains a significant challenge, requiring not only speech recognition but also speaker diarization tasks to be addressed. In this paper, to better address these tasks, we first introduce speaker labels into an autoregressive transformer-based speech recognition model to support multi-speaker overlapped speech recognition. Then, to improve speaker diarization, we propose a novel speaker mask branch to detection the speech segments of individual speakers. With the proposed model, we can perform both speech recognition and speaker diarization tasks simultaneously using a single model. Experimental results on the LibriSpeech-based overlapped dataset demonstrate the effectiveness of the proposed method in both speech recognition and speaker diarization tasks, particularly enhancing the accuracy of speaker diarization in relatively complex multi-talker scenarios.

Fuzzy dark matter (FDM) is an intriguing candidate alternative to the standard cold dark matter (CDM). The FDM model predicts that dark halos have characteristic core structures generated by the effect of quantum pressure, which is different from the structure of CDM halos. We devise a semi-analytic model of a FDM halo density profile by assuming that the density distribution results from the redistribution of mass in a halo with the Navarro-Frenk-White profile. We calculate the mass redistribution radius by considering dynamical relaxation within the FDM halo. We adopt a concentration-halo mass relation with lower concentration compared to that in the CDM model below the half mode mass, which originates from the suppressed matter density fluctuations at small length scales. Our model reproduces the core-halo mass relation (CHMR) found in the numerical simulation of \citet{2014NatPh..10..496S} at $z<1$. We show that the CHMR is well described by a double power law, unlike previous studies that approximate it by a single power law. Our model predictions are in reasonable agreement with the results of the largest FDM simulation of \citet{2021MNRAS.506.2603M} at $z=3$. We find that the core mass for a given halo mass follows the log-normal distribution, both in our model and in the simulation results for the first time, and quantitatively compare the variance of the distribution among them. Although our model does not fully explain the scatter of the CHMR, we show the scatter of the concentration-halo mass relation sizably contributes to them., Comment: 22 pages, 17 figures, published in PRD

Human intelligence relies in part on our brains' ability to create abstract mental models that succinctly capture the hidden blueprint of our reality. Such abstract world models notably allow us to rapidly navigate novel situations by generalizing prior knowledge, a trait deep learning systems have historically struggled to replicate. However, the recent shift from supervised to self-supervised objectives, combined with expressive transformer-based architectures, have yielded powerful foundation models that appear to learn versatile representations that can support a wide range of downstream tasks. This promising development raises the intriguing possibility of such models developing in silico abstract world models. We test this hypothesis by studying the inner workings of small-scale transformers trained to reconstruct partially masked visual scenes generated from a simple blueprint. We show that the network develops intermediate abstract representations, or abstractions, that encode all semantic features of the dataset. These abstractions manifest as low-dimensional manifolds where the embeddings of semantically related tokens transiently converge, thus allowing for the generalization of downstream computations. Using precise manipulation experiments, we demonstrate that abstractions are central to the network's decision-making process. Our research also suggests that these abstractions are compositionally structured, exhibiting features like contextual independence and part-whole relationships that mirror the compositional nature of the dataset. Finally, we introduce a Language-Enhanced Architecture (LEA) designed to encourage the network to articulate its computations. We find that LEA develops an abstraction-centric language that can be easily interpreted, allowing us to more readily access and steer the network's decision-making process.

Astrometric perturbations of critical curves in strong lens systems are thought to be one of the most promising probes of substructures down to small-mass scales. While a smooth mass distribution creates a symmetric geometry of critical curves with radii of curvature about the Einstein radius, substructures introduce small-scale distortions on critical curves, which can break the symmetry of gravitational lensing events near critical curves, such as highly magnified individual stars. We derive a general formula that connects the fluctuation of critical curves with the fluctuation of the surface density caused by substructures, which is useful when constraining models of substructures from observed astrometric perturbations of critical curves. We numerically check that the formula is valid and accurate as long as substructures are not dominated by a small number of massive structures. As a demonstration of the formula, we also explore the possibility that an anomalous position of an extremely magnified star, recently reported as ``Mothra,'' can be explained by fluctuations in the critical curve due to substructures. We find that cold dark matter subhalos with masses ranging from $5 \times 10^7 M_{\odot}/h$ to $10^9 M_{\odot}/h$ can well explain the anomalous position of Mothra, while in the fuzzy dark matter model, the very small mass of $\sim 10^{-24}~\mathrm{eV}$ is needed to explain it., Comment: 12 pages, 5 figures, accepted for publication in Phys. Rev. D

NASA's all-sky survey mission, the Transiting Exoplanet Survey Satellite (TESS), is specifically engineered to detect exoplanets that transit bright stars. Thus far, TESS has successfully identified approximately 400 transiting exoplanets, in addition to roughly 6000 candidate exoplanets pending confirmation. In this study, we present the results of our ongoing project, the Validation of Transiting Exoplanets using Statistical Tools (VaTEST). Our dedicated effort is focused on the confirmation and characterization of new exoplanets through the application of statistical validation tools. Through a combination of ground-based telescope data, high-resolution imaging, and the utilization of the statistical validation tool known as \texttt{TRICERATOPS}, we have successfully discovered eight potential super-Earths. These planets bear the designations: TOI-238b (1.61$^{+0.09} _{-0.10}$ R$_\oplus$), TOI-771b (1.42$^{+0.11} _{-0.09}$ R$_\oplus$), TOI-871b (1.66$^{+0.11} _{-0.11}$ R$_\oplus$), TOI-1467b (1.83$^{+0.16} _{-0.15}$ R$_\oplus$), TOI-1739b (1.69$^{+0.10} _{-0.08}$ R$_\oplus$), TOI-2068b (1.82$^{+0.16} _{-0.15}$ R$_\oplus$), TOI-4559b (1.42$^{+0.13} _{-0.11}$ R$_\oplus$), and TOI-5799b (1.62$^{+0.19} _{-0.13}$ R$_\oplus$). Among all these planets, six of them fall within the region known as 'keystone planets,' which makes them particularly interesting for study. Based on the location of TOI-771b and TOI-4559b below the radius valley we characterized them as likely super-Earths, though radial velocity mass measurements for these planets will provide more details about their characterization. It is noteworthy that planets within the size range investigated herein are absent from our own solar system, making their study crucial for gaining insights into the evolutionary stages between Earth and Neptune., Comment: Accepted: Publications of the Astronomical Society of Australia

Domain shift poses a significant challenge in cross-domain spoken language recognition (SLR) by reducing its effectiveness. Unsupervised domain adaptation (UDA) algorithms have been explored to address domain shifts in SLR without relying on class labels in the target domain. One successful UDA approach focuses on learning domain-invariant representations to align feature distributions between domains. However, disregarding the class structure during the learning process of domain-invariant representations can result in over-alignment, negatively impacting the classification task. To overcome this limitation, we propose an optimal transport (OT)-based UDA algorithm for a cross-domain SLR, leveraging the distribution geometry structure-aware property of OT. An OT-based discrepancy measure on a joint distribution over feature and label information is considered during domain alignment in OT-based UDA. Our previous study discovered that completely aligning the distributions between the source and target domains can introduce a negative transfer, where classes or irrelevant classes from the source domain map to a different class in the target domain during distribution alignment. This negative transfer degrades the performance of the adaptive model. To mitigate this issue, we introduce coupling-weighted partial optimal transport (POT) within our UDA framework for SLR, where soft weighting on the OT coupling based on transport cost is adaptively set during domain alignment. A cross-domain SLR task was used in the experiments to evaluate the proposed UDA. The results demonstrated that our proposed UDA algorithm significantly improved the performance over existing UDA algorithms in a cross-channel SLR task.

Materials following Murray's law are of significant interest due to their unique porous structure and optimal mass transfer ability. However, it is challenging to construct such biomimetic hierarchical channels with perfectly cylindrical pores in synthetic systems following the existing theory. Achieving superior mass transport capacity revealed by Murray's law in nanostructured materials has thus far remained out of reach. We propose a Universal Murray's law applicable to a wide range of hierarchical structures, shapes and generalised transfer processes. We experimentally demonstrate optimal flow of various fluids in hierarchically planar and tubular graphene aerogel structures to validate the proposed law. By adjusting the macroscopic pores in such aerogel-based gas sensors, we also show a significantly improved sensor response dynamic. Our work provides a solid framework for designing synthetic Murray materials with arbitrarily shaped channels for superior mass transfer capabilities, with future implications in catalysis, sensing and energy applications., Comment: 19 pages, 4 figures

Due to the modality discrepancy between textual and acoustic modeling, efficiently transferring linguistic knowledge from a pretrained language model (PLM) to acoustic encoding for automatic speech recognition (ASR) still remains a challenging task. In this study, we propose a cross-modality knowledge transfer (CMKT) learning framework in a temporal connectionist temporal classification (CTC) based ASR system where hierarchical acoustic alignments with the linguistic representation are applied. Additionally, we propose the use of Sinkhorn attention in cross-modality alignment process, where the transformer attention is a special case of this Sinkhorn attention process. The CMKT learning is supposed to compel the acoustic encoder to encode rich linguistic knowledge for ASR. On the AISHELL-1 dataset, with CTC greedy decoding for inference (without using any language model), we achieved state-of-the-art performance with 3.64% and 3.94% character error rates (CERs) for the development and test sets, which corresponding to relative improvements of 34.18% and 34.88% compared to the baseline CTC-ASR system, respectively., Comment: Submitted to ICASSP 2024