Search for: All records

As prompts become central to Large Language Models (LLMs), optimizing them is vital. Textual Stochastic Gradient Descent (TSGD) offers a data-driven approach by iteratively refining prompts using LLM-suggested updates over minibatches. We empirically show that increasing training data initially improves but can later degrade TSGD's performance across NLP tasks, while also raising computational costs. To address this, we propose Textual Stochastic Gradient Descent with Momentum (TSGD-M)—a scalable method that reweights prompt sampling based on past batches. Evaluated on 9 NLP tasks across three domains, TSGD-M outperforms TSGD baselines for most tasks and reduces performance variance. 

We present extensive proper motion measurements of the Crab Nebula made from Canada-France-Hawaii Telescope MegaPrime/MegaCam images taken in 2007, 2016, and 2019. A total of 19974 proper motion vectors with uncertainty < 10 mas yr−1 located over the majority of the Crab Nebula are used to map the supernova remnant’s two-dimensional expansion properties that reflect the dynamics of the original explosion, acceleration of ejecta imparted by spin-down energy from the pulsar, and interaction between the ejecta and surrounding circumstellar material (CSM). The average convergence date we derive is 1105.5 ± 0.5 CE, which is 15-35 years earlier compared to most previous estimates. We find that it varies as a function of position angle around the nebula, with the earliest date and smallest proper motions measured along the equator defined by the east and west bays. The lower acceleration of material along the equatorial plane may be indicative of the supernova’s interaction with a disk-like CSM geometry. Comparing our measurements to previous analytical solutions of the Crab’s expansion and our own numerical simulation using the moving mesh hydrodynamics code Sprout, we conclude that the ejecta have relaxed closer to homologous expansion than expected for the commonly adopted pulsar spindown age of τ ∼ 700 yr and a pulsar wind nebula (PWN) still evolving inside the flat part of the ejecta density profile. These findings provide further evidence that the PWN has broken out of the inner flat part of the supernova ejecta density profile and has experienced “blowout”. 

Predicting the behavior of nanomaterials under various conditions presents a significant challenge due to their complex microstructures. While high-fidelity modeling techniques, such as molecular dynamics (MD) simulations, are effective, they are also computationally demanding. Machine learning (ML) models have opened new avenues for the rapid exploration of design spaces. In this work, we developed a deep learning framework based on a conditional generative adversarial network (cGAN) to predict the evolution of grain boundary (GB) networks in nanocrystalline materials under mechanical loads, incorporating both morphological and atomic details. We conducted MD simulations on nanocrystalline tungsten and used the resulting ground-truth data to train our cGAN model. We assessed the performance of our cGAN model by comparing it to a Convolutional Autoencoder (ConvAE) model and examining the impact of changes in geometric morphology and loading conditions on the model's performance. Our cGAN model demonstrated high accuracy in predicting GB network evolution under a variety of conditions. This developed framework shows potential for predicting various materials' behaviors across a wide range of nanomaterials. 

Abstract We examined three descending positive leaders at distances of 5–11 km and three descending negative leaders at distances of 6–7 km, all simultaneously imaged by high‐speed framing cameras operating in the visible and UV ranges. UV images (290–370 nm) of the positive leaders each exhibited a strong embellishment at the lower channel end, which was not observed in the corresponding visible images (480–800 nm). In contrast, none of the negative leaders exhibited channel embellishment in the UV range and their morphology in UV was similar to that in the visible. Additionally, no embellishment was seen in four negative leaders imaged in UV only. The observed UV embellishment, which is likely to be the streamer zone at the positive‐leader tip, appeared to undergo expansion‐contraction cycles. We attributed the lack of detectable streamer‐zone emission in the UV range in negative leaders to a much lower streamer generation rate compared to positive leaders. 

Abstract We observed five clusters of upper‐level compact intracloud discharges (CIDs) moving positive charge up over land and over water in Florida. The clusters each contained 3 to 6 CIDs, and the overall cluster duration ranged from 27 to 58 s. On average, the CIDs in a given cluster occurred 11 s apart and were separated by a 3D distance of about 1.5 km. All the clustered CIDs were located above the tropopause and were likely associated with convective surges that penetrated the stratosphere. The average periodicity of CID occurrence within a cluster (every 11 s) was comparable to the periodicity at which the average cluster area is expected to be bombarded by ≥10¹⁶ eV cosmic‐ray particles (every 5 s). Each of such energetic particles gives rise to a cosmic ray shower (CRS) and, in the presence of sufficiently strong electric field over a sufficiently large distance, to a relativistic runaway electron avalanche (RREA). We infer that each of our upper‐level CIDs is likely to be caused by a CRS‐RREA traversing, at nearly the speed of light, the electrified overshooting convective surge and triggering, within a few microseconds, a multitude of streamer flashes along its path, over a distance of the order of hundreds of meters (as per the mechanism recently proposed for lightning initiation by Kostinskiy et al., 2020,https://doi.org/10.1029/2020JD033191). The upper‐level CID clustering was likely made possible by the recurring action of energetic cosmic rays and the rapid recovery of the negative screening charge layer at stratospheric altitudes. 

Radio map describes network coverage and is a practically important tool for network planning in modern wireless systems. Generally, radio strength measurements are collected to construct fine-resolution radio maps for analysis. However, certain protected areas are not accessible for measurement due to physical constraints and security considerations, leading to blanked spaces on a radio map. Non-uniformly spaced measurement and uneven observation resolution make it more difficult for radio map estimation and spectrum planning in protected areas. This work explores the distribution of radio spectrum strengths and proposes an exemplar-based approach to reconstruct missing areas on a radio map. Instead of taking generic image processing approaches, we leverage radio propagation models to determine directions of region filling and develop two different schemes to estimate the missing radio signal power. Our test results based on high-fidelity simulation demonstrate efficacy of the proposed methods for radio map reconstruction.