arXiv subjects

Brendan O'Connor

Publications and source records attributed to Brendan O'Connor.

At least 91 records · Page 5Linked to original sources

arXiv2022

ClioQuery: Interactive Query-Oriented Text Analytics for Comprehensive Investigation of Historical News Archives

Historians and archivists often find and analyze the occurrences of query words in newspaper archives, to help answer fundamental questions about society. But much work in text analytics focuses on helping people investigate other textual units, such as events, clusters, ranked documents, entity relationships, or thematic hierarchies. Informed by a study into the needs of historians and archivists, we thus propose ClioQuery, a text analytics system uniquely organized around the analysis of query words in context. ClioQuery applies text simplification techniques from natural language processing to help historians quickly and comprehensively gather and analyze all occurrences of a query word across an archive. It also pairs these new NLP methods with more traditional features like linked views and in-text highlighting to help engender trust in summarization techniques. We evaluate ClioQuery with two separate user studies, in which historians explain how ClioQuery's novel text simplification features can help facilitate historical research. We also evaluate with a separate quantitative comparison study, which shows that ClioQuery helps crowdworkers find and remember historical information. Such results suggest possible new directions for text analytics in other query-oriented settings.

cs.HC↗

arXiv2021

Zero-shot Singing Technique Conversion

In this paper we propose modifications to the neural network framework, AutoVC for the task of singing technique conversion. This includes utilising a pretrained singing technique encoder which extracts technique information, upon which a decoder is conditioned during training. By swapping out a source singer's technique information for that of the target's during conversion, the input spectrogram is reconstructed with the target's technique. We document the beneficial effects of omitting the latent loss, the importance of sequential training, and our process for fine-tuning the bottleneck. We also conducted a listening study where participants rate the specificity of technique-converted voices as well as their naturalness. From this we are able to conclude how effective the technique conversions are and how different conditions affect them, while assessing the model's ability to reconstruct its input data.

cs.SD↗

arXiv2021

Text as Causal Mediators: Research Design for Causal Estimates of Differential Treatment of Social Groups via Language Aspects

Using observed language to understand interpersonal interactions is important in high-stakes decision making. We propose a causal research design for observational (non-experimental) data to estimate the natural direct and indirect effects of social group signals (e.g. race or gender) on speakers' responses with separate aspects of language as causal mediators. We illustrate the promises and challenges of this framework via a theoretical case study of the effect of an advocate's gender on interruptions from justices during U.S. Supreme Court oral arguments. We also discuss challenges conceptualizing and operationalizing causal variables such as gender and language that comprise of many components, and we articulate technical open challenges such as temporal dependence between language mediators in conversational settings.

cs.CL↗

arXiv2021

A month of monitoring the new magnetar Swift J1555.2-5402 during an X-ray outburst

The soft gamma-ray repeater Swift J1555.2-5402 was discovered by means of a 12-ms duration short burst detected with Swift BAT on 2021 June 3. Then 1.6 hours after the first burst detection, NICER started daily monitoring of this X-ray source for a month. The absorbed 2-10 keV flux stays nearly constant at around 4e-11 erg/s/cm2 during the monitoring timespan, showing only a slight gradual decline. A 3.86-s periodicity is detected, and the time derivative of this period is measured to be 3.05(7)e-11 s/s. The soft X-ray pulse shows a single sinusoidal shape with a root-mean-square pulsed fraction that increases as a function of energy from 15% at 1.5 keV to 39% at 7 keV. The equatorial surface magnetic field, characteristic age, and spin-down luminosity are derived under the dipole field approximation to be 3.5e+14 G, 2.0 kyr, and 2.1e+34 erg/s, respectively. An absorbed blackbody with a temperature of 1.1 keV approximates the soft X-ray spectrum. Assuming a source distance of 10 kpc, the peak X-ray luminosity is ~8.5e+35 erg/s in the 2--10 keV band. During the period of observations, we detect 5 and 37 short bursts with Swift/BAT and NICER, respectively. Based on these observational properties, especially the inferred strong magnetic field, this new source is classified as a magnetar. We also coordinated hard X-ray and radio observations with NuSTAR, DSN, and VERA. A hard X-ray power-law component that extends up to at least 40 keV is detected at 3-sigma significance. The 10-60 keV flux, which is dominated by the power-law component, is ~9e-12 erg/s/cm2 with a photon index of ~1.2. The pulsed fraction has a sharp cutoff above 10 keV, down to ~10% in the hard-tail component band. No radio pulsations are detected during the DSN nor VERA observations. We place 7σ upper limits of 0.043mJy and 0.026 mJy on the flux density at S-band and X-band, respectively.

astro-ph.HE↗

arXiv2021

Corpus-Level Evaluation for Event QA: The IndiaPoliceEvents Corpus Covering the 2002 Gujarat Violence

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.

cs.CL↗

arXiv2021

Kilonova Detectability with Wide-Field Instruments

Kilonovae are ultraviolet, optical, and infrared transients powered by the radioactive decay of heavy elements following a neutron star merger. Joint observations of kilonovae and gravitational waves can offer key constraints on the source of Galactic $r$-process enrichment, among other astrophysical topics. However, robust constraints on heavy element production requires rapid kilonova detection (within $\sim 1$ day of merger) as well as multi-wavelength observations across multiple epochs. In this study, we quantify the ability of 13 wide field-of-view instruments to detect kilonovae, leveraging a large grid of over 900 radiative transfer simulations with 54 viewing angles per simulation. We consider both current and upcoming instruments, collectively spanning the full kilonova spectrum. The Roman Space Telescope has the highest redshift reach of any instrument in the study, observing kilonovae out to $z \sim 1$ within the first day post-merger. We demonstrate that BlackGEM, DECam, GOTO, the Vera C. Rubin Observatory's LSST, ULTRASAT, and VISTA can observe some kilonovae out to $z \sim 0.1$ ($\sim$475 Mpc), while DDOTI, MeerLICHT, PRIME, $Swift$/UVOT, and ZTF are confined to more nearby observations. Furthermore, we provide a framework to infer kilonova ejecta properties following non-detections and explore variation in detectability with these ejecta parameters.

astro-ph.HE↗

arXiv2021

Spectrogram Inpainting for Interactive Generation of Instrument Sounds

Modern approaches to sound synthesis using deep neural networks are hard to control, especially when fine-grained conditioning information is not available, hindering their adoption by musicians. In this paper, we cast the generation of individual instrumental notes as an inpainting-based task, introducing novel and unique ways to iteratively shape sounds. To this end, we propose a two-step approach: first, we adapt the VQ-VAE-2 image generation architecture to spectrograms in order to convert real-valued spectrograms into compact discrete codemaps, we then implement token-masked Transformers for the inpainting-based generation of these codemaps. We apply the proposed architecture on the NSynth dataset on masked resampling tasks. Most crucially, we open-source an interactive web interface to transform sounds by inpainting, for artists and practitioners alike, opening up to new, creative uses.

cs.SD↗

arXiv2020

Analyzing Gender Bias within Narrative Tropes

Popular media reflects and reinforces societal biases through the use of tropes, which are narrative elements, such as archetypal characters and plot arcs, that occur frequently across media. In this paper, we specifically investigate gender bias within a large collection of tropes. To enable our study, we crawl tvtropes.org, an online user-created repository that contains 30K tropes associated with 1.9M examples of their occurrences across film, television, and literature. We automatically score the "genderedness" of each trope in our TVTROPES dataset, which enables an analysis of (1) highly-gendered topics within tropes, (2) the relationship between gender bias and popular reception, and (3) how the gender of a work's creator correlates with the types of tropes that they use.

cs.CL↗

arXiv2020

Searching for the radio remnants of short duration gamma-ray bursts

Neutron star mergers produce a substantial amount of fast-moving ejecta, expanding outwardly for years after the merger. The interaction of these ejecta with the surrounding medium may produce a weak isotropic radio remnant, detectable in relatively nearby events. We use late-time radio observations of short duration gamma-ray bursts (sGRBs) to constrain this model. Two samples of events were studied: four sGRBs that are possibly in the local (<200 Mpc) universe were selected to constrain the remnant non-thermal emission from the sub-relativistic ejecta, whereas 17 sGRBs at cosmological distances were used to constrain the presence of a proto-magnetar central engine, possibly re-energezing the merger ejecta. We consider the case of GRB~170817A/GW170817, and find that in this case the early radio emission may be quenched by the jet blast-wave. In all cases, for ejecta mass range of M_ej \lesssim 10^{-2} (5 * 10^{-2}) M_sun, we can rule out very energetic merger ejecta E_ej \gtrsim 5 * 10^{52}(10^{53}) erg, thus excluding the presence of a powerful magnetar as a merger remnant.

astro-ph.HE↗

arXiv2020

Uncertainty over Uncertainty: Investigating the Assumptions, Annotations, and Text Measurements of Economic Policy Uncertainty

Methods and applications are inextricably linked in science, and in particular in the domain of text-as-data. In this paper, we examine one such text-as-data application, an established economic index that measures economic policy uncertainty from keyword occurrences in news. This index, which is shown to correlate with firm investment, employment, and excess market returns, has had substantive impact in both the private sector and academia. Yet, as we revisit and extend the original authors' annotations and text measurements we find interesting text-as-data methodological research questions: (1) Are annotator disagreements a reflection of ambiguity in language? (2) Do alternative text measurements correlate with one another and with measures of external predictive validity? We find for this application (1) some annotator disagreements of economic policy uncertainty can be attributed to ambiguity in language, and (2) switching measurements from keyword-matching to supervised machine learning classifiers results in low correlation, a concerning implication for the validity of the index.

cs.CL↗

arXiv2020

High-cadence observations and variable spin behaviour of magnetar Swift J1818.0-1607 after its outburst

We report on multi-frequency radio observations of the new magnetar Swift J1818.0-1607, following it for more than one month with high cadence. The observations commenced less than 35 hours after its registered first outburst. We obtained timing, polarisation and spectral information. Swift J1818.0-1607 has an unusually steep spectrum for a radio emitting magnetar and also has a relatively narrow and simple pulse profile. The position angle swing of the polarisation is flat over the pulse profile, possibly suggesting that our line-of-sight grazes the edge of the emission beam. This may also explain the steep spectrum. The spin evolution shows large variation in the spin-down rate, associated with four distinct timing events over the course of our observations. Those events may be related to the appearance and disappearance of a second pulse component. The first timing event coincides with our actual observations, while we did not detect significant changes in the emission properties which could reveal further magnetospheric changes. Characteristic ages inferred from the timing measurements over the course of months vary by nearly an order of magnitude. A longer-term spin-down measurement over approximately 100 days suggests an characteristic age of about 500 years, larger than previously reported. Though Swift J1818.0-1607 could still be one of the youngest neutron stars (and magnetars) detected so far, we caution using the characteristic age as a true-age indicator given the caveats behind its calculation.

astro-ph.HE↗

arXiv2020

Constraints on the circumburst environments of short gamma-ray bursts

Observational follow-up of well localized short gamma-ray bursts (SGRBs) has left $20-30\%$ of the population without a coincident host galaxy association to deep optical and NIR limits ($\gtrsim 26$ mag). These SGRBs have been classified as observationally hostless due to their lack of strong host associations. It has been argued that these hostless SGRBs could be an indication of the large distances traversed by the binary neutron star system (due to natal kicks) between its formation and its merger (leading to a SGRB). The distances of GRBs from their host galaxies can be indirectly probed by the surrounding circumburst densities. We show that a lower limit on those densities can be obtained from early afterglow lightcurves. We find that $\lesssim16\%$ of short GRBs in our sample took place at densities $\lesssim10^{-4}$ cm$^{-3}$. These densities represent the expected range of values at distances greater than the host galaxy's virial radii. We find that out of the five SGRBs in our sample that have been found to be observationally hostless, none are consistent with having occurred beyond the virial radius of their birth galaxies. This implies one of two scenarios. Either these observationally hostless SGRBs occurred outside of the half-light radius of their host galaxy, but well within the galactic halo, or in host galaxies at moderate to high redshifts ($z\gtrsim 2$) that were missed by follow-up observations.

astro-ph.HE↗

arXiv2020

Text and Causal Inference: A Review of Using Text to Remove Confounding from Causal Estimates

Many applications of computational social science aim to infer causal conclusions from non-experimental data. Such observational data often contains confounders, variables that influence both potential causes and potential effects. Unmeasured or latent confounders can bias causal estimates, and this has motivated interest in measuring potential confounders from observed text. For example, an individual's entire history of social media posts or the content of a news article could provide a rich measurement of multiple confounders. Yet, methods and applications for this problem are scattered across different communities and evaluation practices are inconsistent. This review is the first to gather and categorize these examples and provide a guide to data-processing and evaluation decisions. Despite increased attention on adjusting for confounding using text, there are still many open problems, which we highlight in this paper.

cs.CL↗

arXiv2019

Investigating Sports Commentator Bias within a Large Corpus of American Football Broadcasts

Sports broadcasters inject drama into play-by-play commentary by building team and player narratives through subjective analyses and anecdotes. Prior studies based on small datasets and manual coding show that such theatrics evince commentator bias in sports broadcasts. To examine this phenomenon, we assemble FOOTBALL, which contains 1,455 broadcast transcripts from American football games across six decades that are automatically annotated with 250K player mentions and linked with racial metadata. We identify major confounding factors for researchers examining racial bias in FOOTBALL, and perform a computational analysis that supports conclusions from prior social science studies.

cs.CL↗

arXiv2019

Query-focused Sentence Compression in Linear Time

Search applications often display shortened sentences which must contain certain query terms and must fit within the space constraints of a user interface. This work introduces a new transition-based sentence compression technique developed for such settings. Our query-focused method constructs length and lexically constrained compressions in linear time, by growing a subgraph in the dependency parse of a sentence. This theoretically efficient approach achieves an 11X empirical speedup over baseline ILP methods, while better reconstructing gold constrained shortenings. Such speedups help query-focused applications, because users are measurably hindered by interface lags. Additionally, our technique does not require an ILP solver or a GPU.

cs.CL↗

arXiv2019

Human acceptability judgements for extractive sentence compression

Recent approaches to English-language sentence compression rely on parallel corpora consisting of sentence-compression pairs. However, a sentence may be shortened in many different ways, which each might be suited to the needs of a particular application. Therefore, in this work, we collect and model crowdsourced judgements of the acceptability of many possible sentence shortenings. We then show how a model of such judgements can be used to support a flexible approach to the compression task. We release our model and dataset for future work.

cs.CL↗

arXiv2018

Evaluating Syntactic Properties of Seq2seq Output with a Broad Coverage HPSG: A Case Study on Machine Translation

Sequence to sequence (seq2seq) models are often employed in settings where the target output is natural language. However, the syntactic properties of the language generated from these models are not well understood. We explore whether such output belongs to a formal and realistic grammar, by employing the English Resource Grammar (ERG), a broad coverage, linguistically precise HPSG-based grammar of English. From a French to English parallel corpus, we analyze the parseability and grammatical constructions occurring in output from a seq2seq translation model. Over 93\% of the model translations are parseable, suggesting that it learns to generate conforming to a grammar. The model has trouble learning the distribution of rarer syntactic rules, and we pinpoint several constructions that differentiate translations between the references and our model.

cs.CL↗

arXiv2018

Monte Carlo Syntax Marginals for Exploring and Using Dependency Parses

Dependency parsing research, which has made significant gains in recent years, typically focuses on improving the accuracy of single-tree predictions. However, ambiguity is inherent to natural language syntax, and communicating such ambiguity is important for error analysis and better-informed downstream applications. In this work, we propose a transition sampling algorithm to sample from the full joint distribution of parse trees defined by a transition-based parsing model, and demonstrate the use of the samples in probabilistic dependency analysis. First, we define the new task of dependency path prediction, inferring syntactic substructures over part of a sentence, and provide the first analysis of performance on this task. Second, we demonstrate the usefulness of our Monte Carlo syntax marginal method for parser error analysis and calibration. Finally, we use this method to propagate parse uncertainty to two downstream information extraction applications: identifying persons killed by police and semantic role assignment.

cs.CL↗