Search for: All records

https://aclanthology.org/2023.acl-short.142/ 

This work reports an approach to print complex patterns of metal nanowires on curvilinear substrates with high conductivity. 

In this paper, we utilize recent advancements in social media natural language processing to obtain state-of-the-art syntactic dependency parsing results for social media English. We observe performance gains of 3.4 UAS and 4.0 LAS against the previous state-of-the-art as well as less disparity between African-American and Mainstream American English dialects. We demonstrate the computational social scientific utility of this parser for the task of socially embedded entity attribute analysis: for a specified entity, derive its semantic relationships from parses’ rich syntax, and accumulate and compare them across social variables. We conduct a case study on politicized views of U.S. official Anthony Fauci during the COVID-19 pandemic. 

We report observations of the optical counterpart of the long gamma-ray burst (GRB) GRB 230812B and its associated supernova (SN) SN 2023pel. The proximity (z= 0.36) and high energy (E_γ,iso∼ 10⁵³erg) make it an important event to study as a probe of the connection between massive star core collapse and relativistic jet formation. With a phenomenological power-law model for the optical afterglow, we find a late-time flattening consistent with the presence of an associated SN. SN 2023pel has an absolute peakr-band magnitude ofM_r= −19.46 ± 0.18 mag (about as bright as SN 1998bw) and evolves on quicker timescales. Using a radioactive heating model, we derive a nickel mass powering the SN ofM_Ni= 0.38 ± 0.01M_⊙and a peak bolometric luminosity ofL_bol∼ 1.3 × 10⁴³erg s⁻¹. We confirm SN 2023pel’s classification as a broad-line Type Ic SN with a spectrum taken 15.5 days after its peak in therband and derive a photospheric expansion velocity ofv_ph= 11,300 ± 1600 km s⁻¹at that phase. Extrapolating this velocity to the time of maximum light, we derive the ejecta massM_ej= 1.0 ± 0.6M_⊙and kinetic energy$E_{\mathrm{KE}}={1.3}_{-1.2}^{+3.3}\times {10}^{51}\mathrm{erg}$. We find that GRB 230812B/SN 2023pel has SN properties that are mostly consistent with the overall GRB-SN population. The lack of correlations found in the GRB-SN population between SN brightness andE_γ,isofor their associated GRBs across a broad range of 7 orders of magnitude provides further evidence that the central engine powering the relativistic ejecta is not coupled to the SN powering mechanism in GRB-SN systems.

 

We report observations of the optical counterpart of the long gamma-ray burst GRB 221009A. Due to the extreme rarity of being both nearby (z= 0.151) and highly energetic (E_γ,iso≥ 10⁵⁴erg), GRB 221009A offers a unique opportunity to probe the connection between massive star core collapse and relativistic jet formation across a very broad range ofγ-ray properties. Adopting a phenomenological power-law model for the afterglow and host galaxy estimates from high-resolution Hubble Space Telescope imaging, we use Bayesian model comparison techniques to determine the likelihood of an associated supernova (SN) contributing excess flux to the optical light curve. Though not conclusive, we find moderate evidence (K_Bayes= 10^1.2) for the presence of an additional component arising from an associated SN, SN 2022xiw, and find that it must be substantially fainter (<67% as bright at the 99% confidence interval) than SN 1998bw. Given the large and uncertain line-of-sight extinction, we attempt to constrain the SN parameters (M_Ni,M_ej, andE_KE) under several different assumptions with respect to the host galaxy’s extinction. We find properties that are broadly consistent with previous GRB-associated SNe:M_Ni= 0.05–0.25M_⊙,M_ej= 3.5–11.1M_⊙, andE_KE= (1.6–5.2) × 10⁵²erg. We note that these properties are weakly constrained due to the faintness of the SN with respect to the afterglow and host emission, but we do find a robust upper limit onM_NiofM_Ni< 0.36M_⊙. Given the tremendous range in isotropic gamma-ray energy release exhibited by GRBs (seven orders of magnitude), the SN emission appears to be decoupled from the central engine in these systems.

 

Automated event extraction in social science applications often requires corpus-level evaluations: for example, aggregating text predictions across metadata and unbiased estimates of recall. We combine corpus-level evaluation requirements with a real-world, social science setting and introduce the IndiaPoliceEvents corpus—all 21,391 sentences from 1,257 English-language Times of India articles about events in the state of Gujarat during March 2002. Our trained annotators read and label every document for mentions of police activity events, allowing for unbiased recall evaluations. In contrast to other datasets with structured event representations, we gather annotations by posing natural questions, and evaluate off-the-shelf models for three different tasks: sentence classification, document ranking, and temporal aggregation of target events. We present baseline results from zero-shot BERT-based models fine-tuned on natural language inference and passage retrieval tasks. Our novel corpus-level evaluations and annotation approach can guide creation of similar social-science-oriented resources in the future.