-
SEC-QA: A Systematic Evaluation Corpus for Financial QA
Authors:
Viet Dac Lai,
Michael Krumdick,
Charles Lovering,
Varshini Reddy,
Craig Schmidt,
Chris Tanner
Abstract:
The financial domain frequently deals with large numbers of long documents that are essential for daily operations. Significant effort is put towards automating financial data analysis. However, a persistent challenge, not limited to the finance domain, is the scarcity of datasets that accurately reflect real-world tasks for model evaluation. Existing datasets are often constrained by size, contex…
▽ More
The financial domain frequently deals with large numbers of long documents that are essential for daily operations. Significant effort is put towards automating financial data analysis. However, a persistent challenge, not limited to the finance domain, is the scarcity of datasets that accurately reflect real-world tasks for model evaluation. Existing datasets are often constrained by size, context, or relevance to practical applications. Moreover, LLMs are currently trained on trillions of tokens of text, limiting access to novel data or documents that models have not encountered during training for unbiased evaluation. We propose SEC-QA, a continuous dataset generation framework with two key features: 1) the semi-automatic generation of Question-Answer (QA) pairs spanning multiple long context financial documents, which better represent real-world financial scenarios; 2) the ability to continually refresh the dataset using the most recent public document collections, not yet ingested by LLMs. Our experiments show that current retrieval augmented generation methods systematically fail to answer these challenging multi-document questions. In response, we introduce a QA system based on program-of-thought that improves the ability to perform complex information retrieval and quantitative reasoning pipelines, thereby increasing QA accuracy.
△ Less
Submitted 20 June, 2024;
originally announced June 2024.
-
Cellular-resolution X-ray microtomography of an entire mouse brain
Authors:
Mattia Humbel,
Christine Tanner,
Marta Girona Alarcón,
Georg Schulz,
Timm Weitkamp,
Mario Scheel,
Vartan Kurtcuoglu,
Bert Müller,
Griffin Rodgers
Abstract:
Purpose: Histology is the gold standard for sub-cellular visualization of the mouse brain. It offers excellent in-plane resolution, but a comparably low out-of-plane resolution due to physical sectioning. X-ray microtomography does not require this trade-off. Tomographic imaging of the entire mouse brain with isotropic cellular resolution produces datasets of multiple terabytes in size. These data…
▽ More
Purpose: Histology is the gold standard for sub-cellular visualization of the mouse brain. It offers excellent in-plane resolution, but a comparably low out-of-plane resolution due to physical sectioning. X-ray microtomography does not require this trade-off. Tomographic imaging of the entire mouse brain with isotropic cellular resolution produces datasets of multiple terabytes in size. These data must be processed and made accessible to domain experts who may have only limited image processing knowledge.
Approach: Extended-field X-ray microtomography covering an entire mouse brain was performed. The 4,495 projections from 8 $\times$ 8 offset acquisitions were stitched to reconstruct a volume of 15,000$^3$ voxels. The microtomography volume was non-rigidly registered to the Allen Mouse Brain Common Coordinate Framework v3 based on a combination of image intensity and landmark pairs.
Results: We present a 3.3 teravoxel dataset covering a full mouse brain with 0.65 $μ$m voxel size. The data were blockwise transformed to a common coordinate system, then stored in a public repository with a hierarchical format for navigation and overlay with anatomical annotations in online viewers such as Neuroglancer or siibra-explorer.
Conclusions: This study demonstrates X-ray imaging and data processing for a full mouse brain, augmenting current atlases by improving resolution in the third dimension by an order of magnitude. The data are publicly available and easily accessible for domain experts via browser-based viewers.
△ Less
Submitted 8 July, 2024; v1 submitted 22 May, 2024;
originally announced May 2024.
-
Aligner-induced tooth movements in three dimensions using clinical data of two patients
Authors:
Ignacio Filippon,
Christine Tanner,
Jeannette A. von Jackowski,
Georg Schulz,
Tino Töpper,
Bert Müller
Abstract:
The performance of optically transparent aligners in orthodontic treatments should be quantified for the individual teeth. To this end, the tooth positions and orientation changes in the three-dimensional space were determined by means of registered, weekly obtained intraoral scans of two patients. The data show the movement and orientation changes of the individual crowns of the upper and lower j…
▽ More
The performance of optically transparent aligners in orthodontic treatments should be quantified for the individual teeth. To this end, the tooth positions and orientation changes in the three-dimensional space were determined by means of registered, weekly obtained intraoral scans of two patients. The data show the movement and orientation changes of the individual crowns of the upper and lower jaws as the result of the forces generated by the series of aligners applied. During the first weeks, the canines and incisors are more affected than the premolars and molars. We detected an overall tooth movement of 1 mm related to a magnitude of extrusion/intrusion of 0.4 mm during a nine week treatment. The data on the orthodontic treatments indicate to what extent the actual tooth movement stays behind the planning represented by the used aligner shapes. The proposed procedure can not only be applied to quantify the clinical outcome of the therapy, but also to improve the planning of the orthodontic treatment for dedicated patients.
△ Less
Submitted 16 May, 2024;
originally announced May 2024.
-
Planet Hunters TESS V: a planetary system around a binary star, including a mini-Neptune in the habitable zone
Authors:
Nora L. Eisner,
Samuel K. Grunblatt,
Oscar Barragán,
Thea H. Faridani,
Chris Lintott,
Suzanne Aigrain,
Cole Johnston,
Ian R. Mason,
Keivan G. Stassun,
Megan Bedell,
Andrew W. Boyle,
David R. Ciardi,
Catherine A. Clark,
Guillaume Hebrard,
David W. Hogg,
Steve B. Howell,
Baptiste Klein,
Joe Llama,
Joshua N. Winn,
Lily L. Zhao,
Joseph M. Akana Murphy,
Corey Beard,
Casey L. Brinkman,
Ashley Chontos,
Pia Cortes-Zuleta
, et al. (39 additional authors not shown)
Abstract:
We report on the discovery and validation of a transiting long-period mini-Neptune orbiting a bright (V = 9.0 mag) G dwarf (TOI 4633; R = 1.05 RSun, M = 1.10 MSun). The planet was identified in data from the Transiting Exoplanet Survey Satellite by citizen scientists taking part in the Planet Hunters TESS project. Modeling of the transit events yields an orbital period of 271.9445 +/- 0.0040 days…
▽ More
We report on the discovery and validation of a transiting long-period mini-Neptune orbiting a bright (V = 9.0 mag) G dwarf (TOI 4633; R = 1.05 RSun, M = 1.10 MSun). The planet was identified in data from the Transiting Exoplanet Survey Satellite by citizen scientists taking part in the Planet Hunters TESS project. Modeling of the transit events yields an orbital period of 271.9445 +/- 0.0040 days and radius of 3.2 +/- 0.20 REarth. The Earth-like orbital period and an incident flux of 1.56 +/- 0.2 places it in the optimistic habitable zone around the star. Doppler spectroscopy of the system allowed us to place an upper mass limit on the transiting planet and revealed a non-transiting planet candidate in the system with a period of 34.15 +/- 0.15 days. Furthermore, the combination of archival data dating back to 1905 with new high angular resolution imaging revealed a stellar companion orbiting the primary star with an orbital period of around 230 years and an eccentricity of about 0.9. The long period of the transiting planet, combined with the high eccentricity and close approach of the companion star makes this a valuable system for testing the formation and stability of planets in binary systems.
△ Less
Submitted 29 April, 2024;
originally announced April 2024.
-
Colloidal dispersions of sterically and electrostatically stabilized PbS quantum dots: the effect of stabilization mechanism on structure factors, second virial coefficients, and film-forming properties
Authors:
Ahhyun Jeong,
Josh Portner,
Christian P. N. Tanner,
Justin C. Ondry,
Chenkun Zhou,
Zehan Mi,
Youssef A. Tazoui,
Vivian R. K. Wall,
Naomi S. Ginsberg,
Dmitri V. Talapin
Abstract:
Electrostatically stabilized nanocrystals (NCs) and, in particular, quantum dots (QDs) hold promise for forming strongly coupled superlattices due to their compact and electronically conductive surface ligands. However, studies on the colloidal dispersion and interparticle interactions of electrostatically stabilized sub-10 nm NCs have been limited, hindering the optimization of colloidal stabilit…
▽ More
Electrostatically stabilized nanocrystals (NCs) and, in particular, quantum dots (QDs) hold promise for forming strongly coupled superlattices due to their compact and electronically conductive surface ligands. However, studies on the colloidal dispersion and interparticle interactions of electrostatically stabilized sub-10 nm NCs have been limited, hindering the optimization of colloidal stability and self-assembly. In this study, we employed small-angle X ray scattering (SAXS) experiments to investigate the interparticle interactions and arrangement of PbS QDs with thiostannate ligands (PbS-Sn2S64-) in polar solvents. The study reveals significant deviations from ideal solution behavior in electrostatically stabilized QD dispersions. Our results demonstrate that PbS-Sn2S64- QDs exhibit long-range interactions within the solvent, in contrast to the short-range steric repulsion characteristic of PbS QDs with oleate ligands (PbS-OA). Introducing highly charged multivalent electrolytes screens electrostatic interactions between charged QDs, reducing the length scale of the repulsive interactions. Furthermore, we calculate the second virial (B2) coefficients from SAXS data, providing insights into how surface chemistry, solvent, and size influence pair potentials. Finally, we explore the influence of long-range interparticle interactions of PbS-Sn2S64- QDs on the morphology of films produced by drying or spin-coating colloidal solutions. The long-range repulsive term of PbS-Sn2S64- QDs promotes the formation of amorphous films, and screening the electrostatic repulsion by addition of an electrolyte enables the formation of a crystalline film. These findings highlight the critical role of NC-NC interactions in tailoring the properties of functional nanomaterials.
△ Less
Submitted 29 April, 2024;
originally announced April 2024.
-
Suppressed self-diffusion of nanoscale constituents of a complex liquid
Authors:
Christian P. N. Tanner,
Vivian R. K. Wall,
Mumtaz Gababa,
Joshua Portner,
Ahhyun Jeong,
Matthew J. Hurley,
Nicholas Leonard,
Jonathan G. Raybin,
James K. Utterback,
Ahyoung Kim,
Andrei Fluerasu,
Yanwen Sun,
Johannes Moeller,
Alexey Zozulya,
Wonhyuk Jo,
Anders Madsen,
Dmitri V. Talapin,
Samuel W. Teitelbaum,
Naomi S. Ginsberg
Abstract:
The ability to understand and ultimately control the transformations and properties of various nanoscale systems, from proteins to synthetic nanomaterial assemblies, hinges on the ability to directly elucidate their dynamics on their characteristic length and time scales. Here, we use MHz X-ray photon correlation spectroscopy (XPCS) to directly elucidate the characteristic microsecond-dynamics of…
▽ More
The ability to understand and ultimately control the transformations and properties of various nanoscale systems, from proteins to synthetic nanomaterial assemblies, hinges on the ability to directly elucidate their dynamics on their characteristic length and time scales. Here, we use MHz X-ray photon correlation spectroscopy (XPCS) to directly elucidate the characteristic microsecond-dynamics of density fluctuations of semiconductor nanocrystals (NCs), not only in a colloidal dispersion but also in a liquid phase consisting of densely packed, yet mobile, NCs with no long-range order. By carefully disentangling X-ray induced effects, we find the wavevector-dependent fluctuation rates in the liquid phase are suppressed relative to those in the colloidal phase and to those in experiments and hydrodynamic theories of densely packed repulsive particles. We show that the suppressed rates are due to a substantial decrease in the self-diffusion of NCs in the liquid phase, which we attribute to explicit attractive interactions. Via comparison with simulations, we find that the extracted strength of the attractions explains the stability of the liquid phase, in contrast to the gelation observed via XPCS in many other charged colloidal systems. This work opens the door to elucidating fast, condensed phase dynamics in a variety of complex fluids and other nanoscale soft matter systems, such as densely packed proteins and non-equilibrium self-assembly processes.
△ Less
Submitted 26 April, 2024;
originally announced April 2024.
-
Enhancing nanocrystal superlattice self-assembly near a metastable liquid binodal
Authors:
Christian P. N. Tanner,
Vivian R. K. Wall,
Joshua Portner,
Ahhyun Jeong,
Avishek Das,
James K. Utterback,
Leo M. Hamerlynck,
Jonathan G. Raybin,
Matthew J. Hurley,
Nicholas Leonard,
Rebecca B. Wai,
Jenna A. Tan,
Mumtaz Gababa,
Chenhui Zhu,
Eric Schaible,
Christopher J. Tassone,
David T. Limmer,
Samuel W. Teitelbaum,
Dmitri V. Talapin,
Naomi S. Ginsberg
Abstract:
Bottom-up assembly of nanocrystals (NCs) into ordered arrays, or superlattices (SLs), is a promising route to design materials with new functionalities, but the degree of control over assembly into functional structures remains challenging. Using electrostatics, rather than density, to tune the interactions between semiconductor NCs, we watch self-assembly proceeding through a metastable liquid ph…
▽ More
Bottom-up assembly of nanocrystals (NCs) into ordered arrays, or superlattices (SLs), is a promising route to design materials with new functionalities, but the degree of control over assembly into functional structures remains challenging. Using electrostatics, rather than density, to tune the interactions between semiconductor NCs, we watch self-assembly proceeding through a metastable liquid phase. We systematically investigate the phase behavior as a function of quench conditions in situ and in real time using small angle X-ray scattering (SAXS). Through quantitative fitting to colloid, liquid, and SL models, we extract the time evolution of each phase and the system phase diagram, which we find to be consistent with short-range attractive interactions. Using the phase diagram's predictive power, we establish control of the self-assembly rate over three orders of magnitude, and identify one- and two-step self-assembly regimes, with only the latter implicating the metastable liquid as an intermediate. Importantly, the presence of the metastable liquid increases SL formation rates relative to the equivalent one-step pathway, and SL order counterintuitively increases with the rate, revealing a highly desirable and generalizable kinetic strategy to promote and enhance ordered assembly.
△ Less
Submitted 25 April, 2024;
originally announced April 2024.
-
Thirteen New M Dwarf + T Dwarf Pairs Identified with WISE/NEOWISE
Authors:
Federico Marocco,
J. Davy Kirkpatrick,
Adam C. Schneider,
Aaron M. Meisner,
Mark Popinchalk,
Christopher R. Gelino,
Jacqueline K. Faherty,
Adam J. Burgasser,
Dan Caselden,
Jonathan Gagné,
Christian Aganze,
Daniella C. Bardalez-Gagliuffi,
Sarah L. Casewell,
Chih-Chun Hsu,
Rocio Kiman,
Peter R. M. Eisenhardt,
Marc J. Kuchner,
Daniel Stern,
Léopold Gramaize,
Arttu Sainio,
Thomas P. Bickle,
Austin Rothermich,
William Pendrill,
Melina Thévenot,
Martin Kabatnik
, et al. (9 additional authors not shown)
Abstract:
We present the discovery of 13 new widely separated T dwarf companions to M dwarf primaries, identified using WISE/NEOWISE data by the CatWISE and Backyard Worlds: Planet 9 projects. This sample represents a $\sim$60% increase in the number of known M+T systems, and allows us to probe the most extreme products of binary/planetary system formation, a discovery space made available by the CatWISE202…
▽ More
We present the discovery of 13 new widely separated T dwarf companions to M dwarf primaries, identified using WISE/NEOWISE data by the CatWISE and Backyard Worlds: Planet 9 projects. This sample represents a $\sim$60% increase in the number of known M+T systems, and allows us to probe the most extreme products of binary/planetary system formation, a discovery space made available by the CatWISE2020 catalog and the Backyard Worlds: Planet 9 effort. Highlights among the sample are WISEP J075108.79-763449.6, a previously known T9 thought to be old due to its SED, which we now find is part of a common-proper-motion pair with L 34-26 A, a well studied young M3 V star within 10 pc of the Sun; CWISE J054129.32-745021.5 B and 2MASS J05581644-4501559 B, two T8 dwarfs possibly associated with the very fast-rotating M4 V stars CWISE J054129.32-745021.5 A and 2MASS J05581644-4501559 A; and UCAC3 52-1038 B, which is among the widest late T companions to main sequence stars, with a projected separation of $\sim$7100 au. The new benchmarks presented here are prime $JWST$ targets, and can help us place strong constraints on formation and evolution theory of substellar objects as well as on atmospheric models for these cold exoplanet analogs.
△ Less
Submitted 22 April, 2024;
originally announced April 2024.
-
Measurement of the ratio of the scalar polarizability to the vector polarizability for the $6s ^2S_{1/2} \rightarrow 7s ^2S_{1/2}$ transition in cesium
Authors:
Jonah A. Quirk,
Carol E. Tanner,
D. S. Elliott
Abstract:
We report measurements of the ratio of the scalar polarizability $α$ to the vector polarizability $β$ for the $6s ^2S_{1/2} \rightarrow 7s ^2S_{1/2}$ transition in atomic cesium. These measurements are motivated by a discrepancy between the values of the vector transition polarizability as determined using two separate methods. In the present measurement, we use a two-pathway, coherent-control tec…
▽ More
We report measurements of the ratio of the scalar polarizability $α$ to the vector polarizability $β$ for the $6s ^2S_{1/2} \rightarrow 7s ^2S_{1/2}$ transition in atomic cesium. These measurements are motivated by a discrepancy between the values of the vector transition polarizability as determined using two separate methods. In the present measurement, we use a two-pathway, coherent-control technique in which we observe the interference between a two-photon interaction driven by infrared light at 1079 nm and a linear Stark-induced interaction driven by the mutually-coherent second harmonic of this infrared beam at 540 nm. The result of our measurements is $α/β= -9.902 \: (9)$, in good agreement with the previous determination of this ratio. This measurement, critical to the study of atomic parity violation in cesium, does not reduce the discrepancy between the two methods for the determination of the vector polarizability $β$ for this transition.
△ Less
Submitted 8 April, 2024;
originally announced April 2024.
-
89 New Ultracool Dwarf Co-Moving Companions Identified With The Backyard Worlds: Planet 9 Citizen Science Project
Authors:
Austin Rothermich,
Jacqueline K. Faherty,
Daniella Bardalez-Gagliuffi,
Adam C. Schneider,
J. Davy Kirkpatrick,
Aaron M. Meisner,
Adam J. Burgasser,
Marc Kuchner,
Katelyn Allers,
Jonathan Gagné,
Dan Caselden,
Emily Calamari,
Mark Popinchalk,
Genaro Suárez,
Roman Gerasimov,
Christian Aganze,
Emma Softich,
Chin-Chun Hsu,
Preethi Karpoor,
Christopher A. Theissen,
Jon Rees,
Rosario Cecilio-Flores-Elie,
Michael C. Cushing,
Federico Marocco,
Sarah Casewell
, et al. (21 additional authors not shown)
Abstract:
We report the identification of 89 new systems containing ultracool dwarf companions to main sequence stars and white dwarfs, using the citizen science project Backyard Worlds: Planet 9 and cross-reference between Gaia and CatWISE2020. Thirty-two of these companions and thirty-three host stars were followed up with spectroscopic observations, with companion spectral types ranging from M7-T9 and ho…
▽ More
We report the identification of 89 new systems containing ultracool dwarf companions to main sequence stars and white dwarfs, using the citizen science project Backyard Worlds: Planet 9 and cross-reference between Gaia and CatWISE2020. Thirty-two of these companions and thirty-three host stars were followed up with spectroscopic observations, with companion spectral types ranging from M7-T9 and host spectral types ranging from G2-M9. These systems exhibit diverse characteristics, from young to old ages, blue to very red spectral morphologies, potential membership to known young moving groups, and evidence of spectral binarity in 9 companions. Twenty of the host stars in our sample show evidence for higher order multiplicity, with an additional 11 host stars being resolved binaries themselves. We compare this sample's characteristics with those of the known stellar binary and exoplanet populations, and find our sample begins to fill in the gap between directly imaged exoplanets and stellary binaries on mass ratio-binding energy plots. With this study, we increase the population of ultracool dwarf companions to FGK stars by $\sim$42\%, and more than triple the known population of ultracool dwarf companions with separations larger than 1,000 au, providing excellent targets for future atmospheric retrievals.
△ Less
Submitted 11 March, 2024; v1 submitted 7 March, 2024;
originally announced March 2024.
-
Greed is All You Need: An Evaluation of Tokenizer Inference Methods
Authors:
Omri Uzan,
Craig W. Schmidt,
Chris Tanner,
Yuval Pinter
Abstract:
While subword tokenizers such as BPE and WordPiece are typically used to build vocabularies for NLP models, the method of decoding text into a sequence of tokens from these vocabularies is often left unspecified, or ill-suited to the method in which they were constructed. We provide a controlled analysis of seven tokenizer inference methods across four different algorithms and three vocabulary siz…
▽ More
While subword tokenizers such as BPE and WordPiece are typically used to build vocabularies for NLP models, the method of decoding text into a sequence of tokens from these vocabularies is often left unspecified, or ill-suited to the method in which they were constructed. We provide a controlled analysis of seven tokenizer inference methods across four different algorithms and three vocabulary sizes, performed on a novel intrinsic evaluation suite we curated for English, combining measures rooted in morphology, cognition, and information theory. We show that for the most commonly used tokenizers, greedy inference performs surprisingly well; and that SaGe, a recently-introduced contextually-informed tokenizer, outperforms all others on morphological alignment.
△ Less
Submitted 31 May, 2024; v1 submitted 2 March, 2024;
originally announced March 2024.
-
Tokenization Is More Than Compression
Authors:
Craig W. Schmidt,
Varshini Reddy,
Haoran Zhang,
Alec Alameddine,
Omri Uzan,
Yuval Pinter,
Chris Tanner
Abstract:
Tokenization is a foundational step in Natural Language Processing (NLP) tasks, bridging raw text and language models. Existing tokenization approaches like Byte-Pair Encoding (BPE) originate from the field of data compression, and it has been suggested that the effectiveness of BPE stems from its ability to condense text into a relatively small number of tokens. We test the hypothesis that fewer…
▽ More
Tokenization is a foundational step in Natural Language Processing (NLP) tasks, bridging raw text and language models. Existing tokenization approaches like Byte-Pair Encoding (BPE) originate from the field of data compression, and it has been suggested that the effectiveness of BPE stems from its ability to condense text into a relatively small number of tokens. We test the hypothesis that fewer tokens lead to better downstream performance by introducing PathPiece, a new tokenizer that segments a document's text into the minimum number of tokens for a given vocabulary. Through extensive experimentation we find this hypothesis not to be the case, casting doubt on the understanding of the reasons for effective tokenization. To examine which other factors play a role, we evaluate design decisions across all three phases of tokenization: pre-tokenization, vocabulary construction, and segmentation, offering new insights into the design of effective tokenizers. Specifically, we illustrate the importance of pre-tokenization and the benefits of using BPE to initialize vocabulary construction. We train 64 language models with varying tokenization, ranging in size from 350M to 2.4B parameters, all of which are made publicly available.
△ Less
Submitted 28 February, 2024;
originally announced February 2024.
-
DocFinQA: A Long-Context Financial Reasoning Dataset
Authors:
Varshini Reddy,
Rik Koncel-Kedziorski,
Viet Dac Lai,
Michael Krumdick,
Charles Lovering,
Chris Tanner
Abstract:
For large language models (LLMs) to be effective in the financial domain -- where each decision can have a significant impact -- it is necessary to investigate realistic tasks and data. Financial professionals often interact with documents that are hundreds of pages long, but most financial research datasets only deal with short excerpts from these documents. To address this, we introduce a long-d…
▽ More
For large language models (LLMs) to be effective in the financial domain -- where each decision can have a significant impact -- it is necessary to investigate realistic tasks and data. Financial professionals often interact with documents that are hundreds of pages long, but most financial research datasets only deal with short excerpts from these documents. To address this, we introduce a long-document financial QA task. We augment 7,437 questions from the existing FinQA dataset with the full-document context, extending the average context length from under 700 words in FinQA to 123k words in DocFinQA. We conduct extensive experiments over retrieval-based QA pipelines and long-context language models. DocFinQA proves a significant challenge for even state-of-the-art systems. We also provide a case-study on the longest documents in DocFinQA and find that models particularly struggle on these documents. Addressing these challenges may have a wide reaching impact across applications where specificity and long-range contexts are critical, like gene sequences and legal document contract analysis.
△ Less
Submitted 29 February, 2024; v1 submitted 12 January, 2024;
originally announced January 2024.
-
In situ coherent X-ray scattering reveals polycrystalline structure and discrete annealing events in strongly-coupled nanocrystal superlattices
Authors:
Matthew J. Hurley,
Christian P. N. Tanner,
Joshua Portner,
James K. Utterback,
Igor Coropceanu,
Garth J. Williams,
Avishek Das,
Andrei Fluerasu,
Yanwen Sun,
Sanghoon Song,
Leo M. Hamerlynck,
Alexander H. Miller,
Priyadarshini Bhattacharyya,
Dmitri V. Talapin,
Naomi S. Ginsberg,
Samuel W. Teitelbaum
Abstract:
Solution-phase bottom up self-assembly of nanocrystals into superstructures such as ordered superlattices is an attractive strategy to generate functional materials of increasing complexity, including very recent advances that incorporate strong interparticle electronic coupling. While the self-assembly kinetics in these systems have been elucidated and related to the product characteristics, the…
▽ More
Solution-phase bottom up self-assembly of nanocrystals into superstructures such as ordered superlattices is an attractive strategy to generate functional materials of increasing complexity, including very recent advances that incorporate strong interparticle electronic coupling. While the self-assembly kinetics in these systems have been elucidated and related to the product characteristics, the weak interparticle bonding interactions suggest the superstructures formed could continue to order within the solution long after the primary nucleation and growth have occurred, even though the mechanism of annealing remains to be elucidated. Here, we use a combination of Bragg coherent diffractive imaging and X-ray photon correlation spectroscopy to create real-space maps of supercrystalline order along with a real-time view of the strain fluctuations in aging strongly coupled nanocrystal superlattices while they remain suspended and immobilized in solution. By combining the results, we deduce that the self-assembled superstructures are polycrystalline, initially comprising multiple nucleation sites, and that shear avalanches at grain boundaries continue to increase crystallinity long after growth has substantially slowed. This multimodal approach should be generalizable to characterize a breadth of materials in situ in their native chemical environments, thus extending the reach of high-resolution coherent X-ray characterization to the benefit of a much wider range of physical systems.
△ Less
Submitted 3 March, 2024; v1 submitted 11 January, 2024;
originally announced January 2024.
-
In situ X-ray scattering reveals coarsening rates of superlattices self-assembled from electrostatically stabilized metal nanocrystals depend non-monotonically on driving force
Authors:
Christian P. N. Tanner,
James K. Utterback,
Joshua Portner,
Igor Coropceanu,
Avishek Das,
Christopher J. Tassone,
Samuel W. Teitelbaum,
David T. Limmer,
Dmitri V. Talapin,
Naomi S. Ginsberg
Abstract:
Self-assembly of colloidal nanocrystals (NCs) into superlattices (SLs) is an appealing strategy to design hierarchically organized materials with new functionalities. Mechanistic studies are still needed to uncover the design principles for SL self-assembly, but such studies have been difficult to perform due to the fast time- and short length scales of NC systems. To address this challenge, we de…
▽ More
Self-assembly of colloidal nanocrystals (NCs) into superlattices (SLs) is an appealing strategy to design hierarchically organized materials with new functionalities. Mechanistic studies are still needed to uncover the design principles for SL self-assembly, but such studies have been difficult to perform due to the fast time- and short length scales of NC systems. To address this challenge, we developed an apparatus to directly measure the evolving phases \textit{in situ} and in real time of an electrostatically stabilized Au NC solution before, during, and after it is quenched to form SLs using small angle X-ray scattering (SAXS). By developing a quantitative model, we fit the time-dependent scattering patterns to obtain the phase diagram of the system and the kinetics of the colloidal and SL phases as a function of varying quench conditions. The extracted phase diagram is consistent with particles whose interactions are short in range relative to their diameter. We find the degree of SL order is primarily determined by fast (sub-second) initial nucleation and growth kinetics, while coarsening at later times depends non-monotonically on the driving force for self-assembly. We validate these results by direct comparison with simulations and use them to suggest dynamic design principles to optimize crystallinity within a finite time window. The combination of this measurement methodology, quantitative analysis, and simulation should be generalizable to elucidate and better control the microscopic self-assembly pathways of a wide range of bottom-up assembled systems and architectures.
△ Less
Submitted 12 February, 2024; v1 submitted 11 December, 2023;
originally announced December 2023.
-
The Initial Mass Function Based on the Full-sky 20-pc Census of $\sim$3,600 Stars and Brown Dwarfs
Authors:
J. Davy Kirkpatrick,
Federico Marocco,
Christopher R. Gelino,
Yadukrishna Raghu,
Jacqueline K. Faherty,
Daniella C. Bardalez Gagliuffi,
Steven D. Schurr,
Kevin Apps,
Adam C. Schneider,
Aaron M. Meisner,
Marc J. Kuchner,
Dan Caselden,
R. L. Smart,
S. L. Casewell,
Roberto Raddi,
Aurora Kesseli,
Nikolaj Stevnbak Andersen,
Edoardo Antonini,
Paul Beaulieu,
Thomas P. Bickle,
Martin Bilsing,
Raymond Chieng,
Guillaume Colin,
Sam Deen,
Alexandru Dereveanco
, et al. (63 additional authors not shown)
Abstract:
A complete accounting of nearby objects -- from the highest-mass white dwarf progenitors down to low-mass brown dwarfs -- is now possible, thanks to an almost complete set of trigonometric parallax determinations from Gaia, ground-based surveys, and Spitzer follow-up. We create a census of objects within a Sun-centered sphere of 20-pc radius and check published literature to decompose each binary…
▽ More
A complete accounting of nearby objects -- from the highest-mass white dwarf progenitors down to low-mass brown dwarfs -- is now possible, thanks to an almost complete set of trigonometric parallax determinations from Gaia, ground-based surveys, and Spitzer follow-up. We create a census of objects within a Sun-centered sphere of 20-pc radius and check published literature to decompose each binary or higher-order system into its separate components. The result is a volume-limited census of $\sim$3,600 individual star formation products useful in measuring the initial mass function across the stellar ($<8 M_\odot$) and substellar ($\gtrsim 5 M_{Jup}$) regimes. Comparing our resulting initial mass function to previous measurements shows good agreement above 0.8$M_\odot$ and a divergence at lower masses. Our 20-pc space densities are best fit with a quadripartite power law, $ξ(M) = dN/dM \propto M^{-α}$ with long-established values of $α= 2.3$ at high masses ($0.55 < M < 8.00 M_\odot$) and $α= 1.3$ at intermediate masses ($0.22 < M < 0.55 M_\odot$), but at lower masses we find $α= 0.25$ for $0.05 < M <0.22 M_\odot$ and $α= 0.6$ for $0.01 < M < 0.05 M_\odot$. This implies that the rate of production as a function of decreasing mass diminishes in the low-mass star/high-mass brown dwarf regime before increasing again in the low-mass brown dwarf regime. Correcting for completeness, we find a star to brown dwarf number ratio of, currently, 4:1, and an average mass per object of 0.41 $M_\odot$.
△ Less
Submitted 6 December, 2023;
originally announced December 2023.
-
Technique for a direct measurement of the cesium anapole moment using coherent rf and Raman interactions
Authors:
Amy Damitz,
Jonah A. Quirk,
Carol E. Tanner,
D. S. Elliott
Abstract:
We report progress toward measurements of the electric dipole (E1) transition moments between hyperfine components of the ground state of atomic cesium. This transition is weakly E1 allowed due to weak interactions between nucleons within the nucleus, which lead to a parity-odd current distribution and its associated anapole moment. In this report, we discuss the experimental geometry of our measu…
▽ More
We report progress toward measurements of the electric dipole (E1) transition moments between hyperfine components of the ground state of atomic cesium. This transition is weakly E1 allowed due to weak interactions between nucleons within the nucleus, which lead to a parity-odd current distribution and its associated anapole moment. In this report, we discuss the experimental geometry of our measurement scheme, explore the effects of extraneous fields that can obscure the signal, present initial measurements, analyze the sources and magnitudes of measurement noise, and suggest improvements to the current apparatus.
△ Less
Submitted 1 February, 2024; v1 submitted 15 November, 2023;
originally announced November 2023.
-
Measurement of the static Stark Shift of the $7s \ ^2S_{1/2}$ level in atomic cesium
Authors:
Jonah A. Quirk,
Aidan Jacobsen,
Amy Damitz,
Carol E. Tanner,
D. S. Elliott
Abstract:
We report a new precision measurement of the dc Stark shift of the $6s\hspace{1mm} ^2S_{1/2} \rightarrow 7s\hspace{1mm}^2S_{1/2}$ transition in atomic cesium-133. Our result is 0.72246 (29) $\textrm{Hz}(\textrm{V}/\textrm{cm})^{-2}$. This result differs from a previous measurement of the Stark shift by $\sim$0.5\%. We use this value to recalculate the magnitude of the reduced dipole matrix element…
▽ More
We report a new precision measurement of the dc Stark shift of the $6s\hspace{1mm} ^2S_{1/2} \rightarrow 7s\hspace{1mm}^2S_{1/2}$ transition in atomic cesium-133. Our result is 0.72246 (29) $\textrm{Hz}(\textrm{V}/\textrm{cm})^{-2}$. This result differs from a previous measurement of the Stark shift by $\sim$0.5\%. We use this value to recalculate the magnitude of the reduced dipole matrix elements $\langle7s ||r||7p_{j}\rangle$, as well as the vector transition polarizability for the $6s \rightarrow 7s$ transition, $\tildeβ = 27.043 \: (36) \ a_0^3$. This determination helps resolve a critical discrepancy between two techniques for determining the vector polarizability.
△ Less
Submitted 19 January, 2024; v1 submitted 15 November, 2023;
originally announced November 2023.
-
BizBench: A Quantitative Reasoning Benchmark for Business and Finance
Authors:
Rik Koncel-Kedziorski,
Michael Krumdick,
Viet Lai,
Varshini Reddy,
Charles Lovering,
Chris Tanner
Abstract:
Answering questions within business and finance requires reasoning, precision, and a wide-breadth of technical knowledge. Together, these requirements make this domain difficult for large language models (LLMs). We introduce BizBench, a benchmark for evaluating models' ability to reason about realistic financial problems. BizBench comprises eight quantitative reasoning tasks, focusing on question-…
▽ More
Answering questions within business and finance requires reasoning, precision, and a wide-breadth of technical knowledge. Together, these requirements make this domain difficult for large language models (LLMs). We introduce BizBench, a benchmark for evaluating models' ability to reason about realistic financial problems. BizBench comprises eight quantitative reasoning tasks, focusing on question-answering (QA) over financial data via program synthesis. We include three financially-themed code-generation tasks from newly collected and augmented QA data. Additionally, we isolate the reasoning capabilities required for financial QA: reading comprehension of financial text and tables for extracting intermediate values, and understanding financial concepts and formulas needed to calculate complex solutions. Collectively, these tasks evaluate a model's financial background knowledge, ability to parse financial documents, and capacity to solve problems with code. We conduct an in-depth evaluation of open-source and commercial LLMs, comparing and contrasting the behavior of code-focused and language-focused models. We demonstrate that the current bottleneck in performance is due to LLMs' limited business and financial understanding, highlighting the value of a challenging benchmark for quantitative reasoning within this domain.
△ Less
Submitted 12 March, 2024; v1 submitted 11 November, 2023;
originally announced November 2023.
-
A Graphical Approach to Document Layout Analysis
Authors:
Jilin Wang,
Michael Krumdick,
Baojia Tong,
Hamima Halim,
Maxim Sokolov,
Vadym Barda,
Delphine Vendryes,
Chris Tanner
Abstract:
Document layout analysis (DLA) is the task of detecting the distinct, semantic content within a document and correctly classifying these items into an appropriate category (e.g., text, title, figure). DLA pipelines enable users to convert documents into structured machine-readable formats that can then be used for many useful downstream tasks. Most existing state-of-the-art (SOTA) DLA models repre…
▽ More
Document layout analysis (DLA) is the task of detecting the distinct, semantic content within a document and correctly classifying these items into an appropriate category (e.g., text, title, figure). DLA pipelines enable users to convert documents into structured machine-readable formats that can then be used for many useful downstream tasks. Most existing state-of-the-art (SOTA) DLA models represent documents as images, discarding the rich metadata available in electronically generated PDFs. Directly leveraging this metadata, we represent each PDF page as a structured graph and frame the DLA problem as a graph segmentation and classification problem. We introduce the Graph-based Layout Analysis Model (GLAM), a lightweight graph neural network competitive with SOTA models on two challenging DLA datasets - while being an order of magnitude smaller than existing models. In particular, the 4-million parameter GLAM model outperforms the leading 140M+ parameter computer vision-based model on 5 of the 11 classes on the DocLayNet dataset. A simple ensemble of these two models achieves a new state-of-the-art on DocLayNet, increasing mAP from 76.8 to 80.8. Overall, GLAM is over 5 times more efficient than SOTA models, making GLAM a favorable engineering choice for DLA tasks.
△ Less
Submitted 3 August, 2023;
originally announced August 2023.
-
What happens before and after: Multi-Event Commonsense in Event Coreference Resolution
Authors:
Sahithya Ravi,
Chris Tanner,
Raymond Ng,
Vered Shwartz
Abstract:
Event coreference models cluster event mentions pertaining to the same real-world event. Recent models rely on contextualized representations to recognize coreference among lexically or contextually similar mentions. However, models typically fail to leverage commonsense inferences, which is particularly limiting for resolving lexically-divergent mentions. We propose a model that extends event men…
▽ More
Event coreference models cluster event mentions pertaining to the same real-world event. Recent models rely on contextualized representations to recognize coreference among lexically or contextually similar mentions. However, models typically fail to leverage commonsense inferences, which is particularly limiting for resolving lexically-divergent mentions. We propose a model that extends event mentions with temporal commonsense inferences. Given a complex sentence with multiple events, e.g., "The man killed his wife and got arrested", with the target event "arrested", our model generates plausible events that happen before the target event - such as "the police arrived", and after it, such as "he was sentenced". We show that incorporating such inferences into an existing event coreference model improves its performance, and we analyze the coreferences in which such temporal knowledge is required.
△ Less
Submitted 21 February, 2023; v1 submitted 19 February, 2023;
originally announced February 2023.
-
Three-dimensional imaging of porcine joint tissues down to the subcellular level
Authors:
Georg Schulz,
Andrea Barbero,
Francine Wolf,
Griffin Rodgers,
Christine Tanner,
Timm Weitkamp,
Marcus Mumme,
Marta Morawska,
Daniel Beer,
Bert Müller
Abstract:
Joint tissues consist of trabecular and cortical bone as well as calcified and hyaline cartilage, which presents a challenge for hard X-ray-based visualization on the sub-cellular level due to the wide range of local X-ray absorption values. The density of the calcified tissues requires rather high photon energy, which often leads to insufficient contrast within the cartilage and impedes the visua…
▽ More
Joint tissues consist of trabecular and cortical bone as well as calcified and hyaline cartilage, which presents a challenge for hard X-ray-based visualization on the sub-cellular level due to the wide range of local X-ray absorption values. The density of the calcified tissues requires rather high photon energy, which often leads to insufficient contrast within the cartilage and impedes the visualization of individual biological cells. Decalcification of the tissues reduces the total and local X-ray absorption values and allows for selecting a lower photon energy. Further contrast enhancement can be achieved by ethanol fixation and paraffin tissue embedding. In this study, we (i) searched for an appropriate visualization method to investigate lesions generated by a laser osteotome and (ii) visualized a decalcified porcine joint after ethanol fixation and subsequent paraffin embedding using laboratory- and synchrotron radiation-based microtomography. The experiments at the ANATOMIX beamline of Synchrotron SOLEIL were performed in off-axis scan mode with a pixel size of 1.3 um. Individual cells in all layers of the joint could be made visible and the effect of ethanol fixation and paraffin embedding demonstrated.
△ Less
Submitted 17 November, 2022; v1 submitted 16 November, 2022;
originally announced November 2022.
-
Measurement of the hyperfine coupling constants and absolute energies of the $8p \ ^2P_{1/2}$ and $8p \ ^2P_{3/2}$ levels in atomic cesium
Authors:
Jonah Quirk,
Liam Sherman,
Amy Damitz,
Carol Tanner,
D. S. Elliott
Abstract:
We report measurements of the hyperfine coupling constant for the $8p \ ^2P_{1/2}$ level of atomic cesium, $^{133}$Cs, with a relative uncertainty of $\sim$0.019\%. Our result is $A = 42.933 \: (8)$ MHz, in good agreement with recent theoretical results. We also examine the hyperfine structure of the $8p \ ^2P_{3/2}$ state, and derive new values for the state energies of the $8p \ ^2P_{1/2}$ and…
▽ More
We report measurements of the hyperfine coupling constant for the $8p \ ^2P_{1/2}$ level of atomic cesium, $^{133}$Cs, with a relative uncertainty of $\sim$0.019\%. Our result is $A = 42.933 \: (8)$ MHz, in good agreement with recent theoretical results. We also examine the hyperfine structure of the $8p \ ^2P_{3/2}$ state, and derive new values for the state energies of the $8p \ ^2P_{1/2}$ and $8p \ ^2P_{3/2}$ states of cesium.
△ Less
Submitted 16 September, 2022;
originally announced September 2022.
-
LineCap: Line Charts for Data Visualization Captioning Models
Authors:
Anita Mahinpei,
Zona Kostic,
Chris Tanner
Abstract:
Data visualization captions help readers understand the purpose of a visualization and are crucial for individuals with visual impairments. The prevalence of poor figure captions and the successful application of deep learning approaches to image captioning motivate the use of similar techniques for automated figure captioning. However, research in this field has been stunted by the lack of suitab…
▽ More
Data visualization captions help readers understand the purpose of a visualization and are crucial for individuals with visual impairments. The prevalence of poor figure captions and the successful application of deep learning approaches to image captioning motivate the use of similar techniques for automated figure captioning. However, research in this field has been stunted by the lack of suitable datasets. We introduce LineCap, a novel figure captioning dataset of 3,528 figures, and we provide insights from curating this dataset and using end-to-end deep learning models for automated figure captioning.
△ Less
Submitted 14 July, 2022;
originally announced July 2022.
-
What GPT Knows About Who is Who
Authors:
Xiaohan Yang,
Eduardo Peynetti,
Vasco Meerman,
Chris Tanner
Abstract:
Coreference resolution -- which is a crucial task for understanding discourse and language at large -- has yet to witness widespread benefits from large language models (LLMs). Moreover, coreference resolution systems largely rely on supervised labels, which are highly expensive and difficult to annotate, thus making it ripe for prompt engineering. In this paper, we introduce a QA-based prompt-eng…
▽ More
Coreference resolution -- which is a crucial task for understanding discourse and language at large -- has yet to witness widespread benefits from large language models (LLMs). Moreover, coreference resolution systems largely rely on supervised labels, which are highly expensive and difficult to annotate, thus making it ripe for prompt engineering. In this paper, we introduce a QA-based prompt-engineering method and discern \textit{generative}, pre-trained LLMs' abilities and limitations toward the task of coreference resolution. Our experiments show that GPT-2 and GPT-Neo can return valid answers, but that their capabilities to identify coreferent mentions are limited and prompt-sensitive, leading to inconsistent results.
△ Less
Submitted 15 May, 2022;
originally announced May 2022.
-
Evaluating the Fairness Impact of Differentially Private Synthetic Data
Authors:
Blake Bullwinkel,
Kristen Grabarz,
Lily Ke,
Scarlett Gong,
Chris Tanner,
Joshua Allen
Abstract:
Differentially private (DP) synthetic data is a promising approach to maximizing the utility of data containing sensitive information. Due to the suppression of underrepresented classes that is often required to achieve privacy, however, it may be in conflict with fairness. We evaluate four DP synthesizers and present empirical results indicating that three of these models frequently degrade fairn…
▽ More
Differentially private (DP) synthetic data is a promising approach to maximizing the utility of data containing sensitive information. Due to the suppression of underrepresented classes that is often required to achieve privacy, however, it may be in conflict with fairness. We evaluate four DP synthesizers and present empirical results indicating that three of these models frequently degrade fairness outcomes on downstream binary classification tasks. We draw a connection between fairness and the proportion of minority groups present in the generated synthetic data, and find that training synthesizers on data that are pre-processed via a multi-label undersampling method can promote more fair outcomes without degrading accuracy.
△ Less
Submitted 20 June, 2022; v1 submitted 9 May, 2022;
originally announced May 2022.
-
Three-dimensional imaging and analysis of annual layers in tree trunk and tooth cementum
Authors:
Bert Müller,
Muriel Stiefel,
Griffin Rodgers,
Mattia Humbel,
Melissa Osterwalder,
Jeannette von Jackowski,
Gerhard Hotz,
Adriana A. Velasco,
Henry T. Bunn,
Mario Scheel,
Timm Weitkamp,
Georg Schulz,
Christine Tanner
Abstract:
The growth of plants, animals, and humans can give rise to layered structures associated with annual periodicity. Thickness variations are often correlated to nutrition supply and stress factors. The annual layers in a tree trunk with millimeter thickness can be directly counted, whereas the layers in tooth cementum with micrometer thickness are made visible using optical microscopy. These optical…
▽ More
The growth of plants, animals, and humans can give rise to layered structures associated with annual periodicity. Thickness variations are often correlated to nutrition supply and stress factors. The annual layers in a tree trunk with millimeter thickness can be directly counted, whereas the layers in tooth cementum with micrometer thickness are made visible using optical microscopy. These optical techniques rely on the surface evaluation or thin, optically transparent slices. Hard X-ray tomography with micrometer resolution, however, provides a three-dimensional view without physical slicing. We have developed a procedure to enhance the tomography data of annual layers in human and bovid tooth cementum. The analysis of a substantial part of an archeological human tooth demonstrated that the detected number of layers depended on the selected region and could vary between 13 and 27. The related average thickness of the annual layers was found to be (5.4 $\pm$ 1.9) $μ$m for the human tooth, whereas the buffalo tooth exhibited a layer periodicity of 46 $μ$m. The present study elucidates the potential of combining computational tools with high-quality micro computed tomography data to quantify the annual layers in tooth cementum for a variety of purposes including age-at-death determination.
△ Less
Submitted 25 April, 2022;
originally announced May 2022.
-
Automatic Fake News Detection: Are current models "fact-checking" or "gut-checking"?
Authors:
Ian Kelk,
Benjamin Basseri,
Wee Yi Lee,
Richard Qiu,
Chris Tanner
Abstract:
Automatic fake news detection models are ostensibly based on logic, where the truth of a claim made in a headline can be determined by supporting or refuting evidence found in a resulting web query. These models are believed to be reasoning in some way; however, it has been shown that these same results, or better, can be achieved without considering the claim at all -- only the evidence. This imp…
▽ More
Automatic fake news detection models are ostensibly based on logic, where the truth of a claim made in a headline can be determined by supporting or refuting evidence found in a resulting web query. These models are believed to be reasoning in some way; however, it has been shown that these same results, or better, can be achieved without considering the claim at all -- only the evidence. This implies that other signals are contained within the examined evidence, and could be based on manipulable factors such as emotion, sentiment, or part-of-speech (POS) frequencies, which are vulnerable to adversarial inputs. We neutralize some of these signals through multiple forms of both neural and non-neural pre-processing and style transfer, and find that this flattening of extraneous indicators can induce the models to actually require both claims and evidence to perform well. We conclude with the construction of a model using emotion vectors built off a lexicon and passed through an "emotional attention" mechanism to appropriately weight certain emotions. We provide quantifiable results that prove our hypothesis that manipulable features are being used for fact-checking.
△ Less
Submitted 14 April, 2022;
originally announced April 2022.
-
Substellar Hyades Candidates from the UKIRT Hemisphere Survey
Authors:
Adam C. Schneider,
Frederick J. Vrba,
Jeffrey A. Munn,
Scott E. Dahm,
Justice Bruursema,
Stephen J. Williams,
Byran N. Dorland,
Jacqueline K. Faherty,
Austin Rothermich,
Emily Calamari,
Michael C. Cushing,
Dan Caselden,
Martin Kabatnik,
William Pendrill,
Arttu Sainio,
Nikolaj Stevnbak Andersen,
Christopher Tanner
Abstract:
We have used data from the UKIRT Hemisphere Survey (UHS) to search for substellar members of the Hyades cluster. Our search recovered several known substellar Hyades members, and two known brown dwarfs that we suggest may be members based on a new kinematic analysis. We uncovered thirteen new substellar Hyades candidates, and obtained near-infrared follow-up spectroscopy of each with IRTF/SpeX. Si…
▽ More
We have used data from the UKIRT Hemisphere Survey (UHS) to search for substellar members of the Hyades cluster. Our search recovered several known substellar Hyades members, and two known brown dwarfs that we suggest may be members based on a new kinematic analysis. We uncovered thirteen new substellar Hyades candidates, and obtained near-infrared follow-up spectroscopy of each with IRTF/SpeX. Six candidates with spectral types between M7 and L0 are ruled out as potential members based on their photometric distances ($\gtrsim$100 pc). The remaining seven candidates, with spectral types between L5 and T4, are all potential Hyades members, with five showing strong membership probabilities based on BANYAN $Σ$ and a convergent point analysis. Distances and radial velocities are still needed to confirm Hyades membership. If confirmed, these would be some of the lowest mass free-floating members of the Hyades yet known, with masses as low as $\sim$30 $M_{\rm Jup}$. An analysis of all known substellar Hyades candidates shows evidence that the full extent of the Hyades has yet to be probed for low-mass members, and more would likely be recovered with deeper photometric and astrometric investigations.
△ Less
Submitted 21 March, 2022;
originally announced March 2022.
-
Planet Hunters TESS IV: A massive, compact hierarchical triple star system TIC 470710327
Authors:
Nora L. Eisner,
Cole Johnston,
Silvia Toonen,
Abigail J. Frost,
Soetkin Janssens,
Chris J. Lintott,
Suzanne Aigrain,
Hugues Sana,
Michael Abdul-Masih,
Karla Z. Arellano-Córdova,
Paul G. Beck,
Emma Bordier,
Emily Canon,
Ana Escorza,
Mattias Fabry,
Lars Hermansson,
Steve Howell,
Grant Miller,
Shreeya Sheyte,
Safaa Alhassan,
Elisabeth M. L. Baeten,
Frank Barnet,
Stewart. J. Bean,
Mikael Bernau,
David M. Bundy
, et al. (15 additional authors not shown)
Abstract:
We report the discovery and analysis of a massive, compact, hierarchical triple system (TIC 470710327) initially identified by citizen scientists in data obtained by NASA's Transiting Exoplanet Survey Satellite (TESS). Spectroscopic follow-up observations obtained with the HERMES spectrograph, combined with eclipse timing variations (ETVs), confirm that the system is comprised of three OB stars, w…
▽ More
We report the discovery and analysis of a massive, compact, hierarchical triple system (TIC 470710327) initially identified by citizen scientists in data obtained by NASA's Transiting Exoplanet Survey Satellite (TESS). Spectroscopic follow-up observations obtained with the HERMES spectrograph, combined with eclipse timing variations (ETVs), confirm that the system is comprised of three OB stars, with a compact 1.10 d eclipsing binary and a non-eclipsing tertiary on a 52.04 d orbit. Dynamical modelling of the system (from radial velocity and ETVs) reveal a rare configuration wherein the tertiary star (O9.5-B0.5V; 14-17 M$_{\odot}$) is more massive than the combined mass of the inner binary (10.9-13.2 M$_{\odot}$). Given the high mass of the tertiary, we predict that this system will undergo multiple phases of mass transfer in the future, and likely end up as a double neutron star gravitational wave progenitor or an exotic Thorne-Zytkow object. Further observational characterisation of this system promises constraints on both formation scenarios of massive stars as well as their exotic evolutionary end-products.
△ Less
Submitted 14 February, 2022;
originally announced February 2022.
-
Measurement of the hyperfine coupling constants and absolute energies of the $12s \ ^2S_{1/2}$, $13s \ ^2S_{1/2}$, and $11d \ ^2D_{J}$ levels in atomic cesium
Authors:
Jonah A. Quirk,
Amy Damitz,
Carol E. Tanner,
D. S. Elliott
Abstract:
We report measurements of the absolute energies of the hyperfine components of the $12s \ ^2S_{1/2}$ and $13s \ ^2S_{1/2}$ levels of atomic cesium, $^{133}$Cs. Using the frequency difference between these components, we determine the hyperfine coupling constants for these states, and report these values with a relative uncertainty of $\sim$0.06 %. We also examine the hyperfine structure of the…
▽ More
We report measurements of the absolute energies of the hyperfine components of the $12s \ ^2S_{1/2}$ and $13s \ ^2S_{1/2}$ levels of atomic cesium, $^{133}$Cs. Using the frequency difference between these components, we determine the hyperfine coupling constants for these states, and report these values with a relative uncertainty of $\sim$0.06 %. We also examine the hyperfine structure of the $11d \ ^2D_{J}$ ($J=3/2, 5/2$) states, and resolve the sign ambiguity of the hyperfine coupling constants from previous measurements of these states. We also derive new, high precision values for the state energies of the $12s \ ^2S_{1/2}$, $13s \ ^2S_{1/2}$ and $11d \ ^2D_{J}$ states of cesium.
△ Less
Submitted 3 January, 2022;
originally announced January 2022.
-
Discovery of 16 New Members of the Solar Neighborhood using Proper Motions from CatWISE2020
Authors:
Tarun Kota,
J. Davy Kirkpatrick,
Dan Caselden,
Federico Marocco,
Adam C. Schneider,
Jonathan Gagné,
Jacqueline K. Faherty,
Aaron M. Meisner,
Marc J. Kuchner,
Sarah Casewell,
Kanishk Kacholia,
Tom Bickle,
Paul Beaulieu,
Guillaume Colin,
Leslie K. Hamlet,
Jörg Schümann,
Christopher Tanner,
the Backyard Worlds,
:,
Planet 9 Collaboration
Abstract:
In an effort to identify nearby and unusual cold objects in the solar neighborhood, we searched for previously unidentified moving objects using CatWISE2020 proper motion data combined with machine learning methods. We paired the motion candidates with their counterparts in 2MASS, UHS, and VHS. Then we searched for white dwarf, brown dwarf, and subdwarf outliers on the resulting color-color diagra…
▽ More
In an effort to identify nearby and unusual cold objects in the solar neighborhood, we searched for previously unidentified moving objects using CatWISE2020 proper motion data combined with machine learning methods. We paired the motion candidates with their counterparts in 2MASS, UHS, and VHS. Then we searched for white dwarf, brown dwarf, and subdwarf outliers on the resulting color-color diagrams. This resulted in the discovery of 16 new dwarfs including two nearby M dwarfs (< 30 pc), a possible young L dwarf, a high motion early T dwarf and 3 later T dwarfs. This research represents a step forward in completing the census of the Sun's neighbors.
△ Less
Submitted 22 December, 2021; v1 submitted 16 December, 2021;
originally announced December 2021.
-
Building astroBERT, a language model for Astronomy & Astrophysics
Authors:
Felix Grezes,
Sergi Blanco-Cuaresma,
Alberto Accomazzi,
Michael J. Kurtz,
Golnaz Shapurian,
Edwin Henneken,
Carolyn S. Grant,
Donna M. Thompson,
Roman Chyla,
Stephen McDonald,
Timothy W. Hostetler,
Matthew R. Templeton,
Kelly E. Lockhart,
Nemanja Martinovic,
Shinyi Chen,
Chris Tanner,
Pavlos Protopapas
Abstract:
The existing search tools for exploring the NASA Astrophysics Data System (ADS) can be quite rich and empowering (e.g., similar and trending operators), but researchers are not yet allowed to fully leverage semantic search. For example, a query for "results from the Planck mission" should be able to distinguish between all the various meanings of Planck (person, mission, constant, institutions and…
▽ More
The existing search tools for exploring the NASA Astrophysics Data System (ADS) can be quite rich and empowering (e.g., similar and trending operators), but researchers are not yet allowed to fully leverage semantic search. For example, a query for "results from the Planck mission" should be able to distinguish between all the various meanings of Planck (person, mission, constant, institutions and more) without further clarification from the user. At ADS, we are applying modern machine learning and natural language processing techniques to our dataset of recent astronomy publications to train astroBERT, a deeply contextual language model based on research at Google. Using astroBERT, we aim to enrich the ADS dataset and improve its discoverability, and in particular we are developing our own named entity recognition tool. We present here our preliminary results and lessons learned.
△ Less
Submitted 1 December, 2021;
originally announced December 2021.
-
Modularity maximization as a flexible and generic framework for brain network exploratory analysis
Authors:
Farnaz Zamani Esfahlani,
Youngheun Jo,
Maria Grazia Puxeddu,
Haily Merritt,
Jacob C. Tanner,
Sarah Greenwell,
Riya Patel,
Joshua Faskowitz,
Richard F. Betzel
Abstract:
The modular structure of brain networks supports specialized information processing, complex dynamics, and cost-efficient spatial embedding. Inter-individual variation in modular structure has been linked to differences in performance, disease, and development. There exist many data-driven methods for detecting and comparing modular structure, the most popular of which is modularity maximization.…
▽ More
The modular structure of brain networks supports specialized information processing, complex dynamics, and cost-efficient spatial embedding. Inter-individual variation in modular structure has been linked to differences in performance, disease, and development. There exist many data-driven methods for detecting and comparing modular structure, the most popular of which is modularity maximization. Although modularity maximization is a general framework that can be modified and reparamaterized to address domain-specific research questions, its application to neuroscientific datasets has, thus far, been narrow. Here, we highlight several strategies in which the ``out-of-the-box'' version of modularity maximization can be extended to address questions specific to neuroscience. First, we present approaches for detecting ``space-independent'' modules and for applying modularity maximization to signed matrices. Next, we show that the modularity maximization frame is well-suited for detecting task- and condition-specific modules. Finally, we highlight the role of multi-layer models in detecting and tracking modules across time, tasks, subjects, and modalities. In summary, modularity maximization is a flexible and general framework that can be adapted to detect modular structure resulting from a wide range of hypotheses. This article highlights opens multiple frontiers for future research and applications.
△ Less
Submitted 29 June, 2021;
originally announced June 2021.
-
The Field Substellar Mass Function Based on the Full-sky 20-pc Census of 525 L, T, and Y Dwarfs
Authors:
J. Davy Kirkpatrick,
Christopher R. Gelino,
Jacqueline K. Faherty,
Aaron M. Meisner,
Dan Caselden,
Adam C. Schneider,
Federico Marocco,
Alfred J. Cayago,
R. L. Smart,
Peter R. Eisenhardt,
Marc J. Kuchner,
Edward L. Wright,
Michael C. Cushing,
Katelyn N. Allers,
Daniella C. Bardalez Gagliuffi,
Adam J. Burgasser,
Jonathan Gagne,
Sarah E. Logsdon,
Emily C. Martin,
James G. Ingalls,
Patrick J. Lowrance,
Ellianna S. Abrahams,
Christian Aganze,
Roman Gerasimov,
Eileen C. Gonzales
, et al. (27 additional authors not shown)
Abstract:
We present final Spitzer trigonometric parallaxes for 361 L, T, and Y dwarfs. We combine these with prior studies to build a list of 525 known L, T, and Y dwarfs within 20 pc of the Sun, 38 of which are presented here for the first time. Using published photometry and spectroscopy as well as our own follow-up, we present an array of color-magnitude and color-color diagrams to further characterize…
▽ More
We present final Spitzer trigonometric parallaxes for 361 L, T, and Y dwarfs. We combine these with prior studies to build a list of 525 known L, T, and Y dwarfs within 20 pc of the Sun, 38 of which are presented here for the first time. Using published photometry and spectroscopy as well as our own follow-up, we present an array of color-magnitude and color-color diagrams to further characterize census members, and we provide polynomial fits to the bulk trends. Using these characterizations, we assign each object a $T_{\rm eff}$ value and judge sample completeness over bins of $T_{\rm eff}$ and spectral type. Except for types $\ge$ T8 and $T_{\rm eff} <$ 600K, our census is statistically complete to the 20-pc limit. We compare our measured space densities to simulated density distributions and find that the best fit is a power law ($dN/dM \propto M^{-α}$) with $α= 0.6{\pm}0.1$. We find that the evolutionary models of Saumon & Marley correctly predict the observed magnitude of the space density spike seen at 1200K $< T_{\rm eff} <$ 1350K, believed to be caused by an increase in the cooling timescale across the L/T transition. Defining the low-mass terminus using this sample requires a more statistically robust and complete sample of dwarfs $\ge$Y0.5 and with $T_{\rm eff} <$ 400K. We conclude that such frigid objects must exist in substantial numbers, despite the fact that few have so far been identified, and we discuss possible reasons why they have largely eluded detection.
△ Less
Submitted 23 November, 2020;
originally announced November 2020.
-
Spitzer Follow-up of Extremely Cold Brown Dwarfs Discovered by the Backyard Worlds: Planet 9 Citizen Science Project
Authors:
Aaron M. Meisner,
Jacqueline K. Faherty,
J. Davy Kirkpatrick,
Adam C. Schneider,
Dan Caselden,
Jonathan Gagne,
Marc J. Kuchner,
Adam J. Burgasser,
Sarah L. Casewell,
John H. Debes,
Etienne Artigau,
Daniella C. Bardalez Gagliuffi,
Sarah E. Logsdon,
Rocio Kiman,
Katelyn Allers,
Chih-Chun Hsu,
John P. Wisniewski,
Michaela B. Allen,
Paul Beaulieu,
Guillaume Colin,
Hugo A. Durantini Luca,
Sam Goodman,
Leopold Gramaize,
Leslie K. Hamlet,
Ken Hinckley
, et al. (18 additional authors not shown)
Abstract:
We present Spitzer follow-up imaging of 95 candidate extremely cold brown dwarfs discovered by the Backyard Worlds: Planet 9 citizen science project, which uses visually perceived motion in multi-epoch WISE images to identify previously unrecognized substellar neighbors to the Sun. We measure Spitzer [3.6]-[4.5] color to phototype our brown dwarf candidates, with an emphasis on pinpointing the col…
▽ More
We present Spitzer follow-up imaging of 95 candidate extremely cold brown dwarfs discovered by the Backyard Worlds: Planet 9 citizen science project, which uses visually perceived motion in multi-epoch WISE images to identify previously unrecognized substellar neighbors to the Sun. We measure Spitzer [3.6]-[4.5] color to phototype our brown dwarf candidates, with an emphasis on pinpointing the coldest and closest Y dwarfs within our sample. The combination of WISE and Spitzer astrometry provides quantitative confirmation of the transverse motion of 75 of our discoveries. Nine of our motion-confirmed objects have best-fit linear motions larger than 1"/yr; our fastest-moving discovery is WISEA J155349.96+693355.2 (total motion ~2.15"/yr), a possible T type subdwarf. We also report a newly discovered wide-separation (~400 AU) T8 comoving companion to the white dwarf LSPM J0055+5948 (the fourth such system to be found), plus a candidate late T companion to the white dwarf LSR J0002+6357 at 5.5' projected separation (~8,700 AU if associated). Among our motion-confirmed targets, five have Spitzer colors most consistent with spectral type Y. Four of these five have exceptionally red Spitzer colors suggesting types of Y1 or later, adding considerably to the small sample of known objects in this especially valuable low-temperature regime. Our Y dwarf candidates begin bridging the gap between the bulk of the Y dwarf population and the coldest known brown dwarf.
△ Less
Submitted 14 August, 2020;
originally announced August 2020.
-
Infinite-order perturbative treatment for quantum evolution with exchange
Authors:
Jacob R. Lindale,
Shannon L. Eriksson,
Christian P. N. Tanner,
Warren S. Warren
Abstract:
Many important applications in biochemistry, materials science, and catalysis sit squarely at the interface between quantum and statistical mechanics: coherent evolution is interrupted by discrete events, such as binding of a substrate or isomerization. Theoretical models for such dynamics usually truncate the incorporation of these events to the linear-response limit, thus requiring small step si…
▽ More
Many important applications in biochemistry, materials science, and catalysis sit squarely at the interface between quantum and statistical mechanics: coherent evolution is interrupted by discrete events, such as binding of a substrate or isomerization. Theoretical models for such dynamics usually truncate the incorporation of these events to the linear-response limit, thus requiring small step sizes. Here, we completely re-assess the foundations of chemical exchange models and redesign a master equation treatment accurate to all orders in perturbation theory. The net result is an astonishingly simple correction to the traditional picture which vastly improves convergence with no increased computational cost. We demonstrate that this approach accurately and efficiently extracts physical parameters from highly complex experimental data, such as coherent hyperpolarization dynamics in magnetic resonance, and is applicable to a wide range of other systems.
△ Less
Submitted 4 March, 2020;
originally announced March 2020.
-
Delineating Bone Surfaces in B-Mode Images Constrained by Physics of Ultrasound Propagation
Authors:
Firat Ozdemir,
Christine Tanner,
Orcun Goksel
Abstract:
Bone surface delineation in ultrasound is of interest due to its potential in diagnosis, surgical planning, and post-operative follow-up in orthopedics, as well as the potential of using bones as anatomical landmarks in surgical navigation. We herein propose a method to encode the physics of ultrasound propagation into a factor graph formulation for the purpose of bone surface delineation. In this…
▽ More
Bone surface delineation in ultrasound is of interest due to its potential in diagnosis, surgical planning, and post-operative follow-up in orthopedics, as well as the potential of using bones as anatomical landmarks in surgical navigation. We herein propose a method to encode the physics of ultrasound propagation into a factor graph formulation for the purpose of bone surface delineation. In this graph structure, unary node potentials encode the local likelihood for being a soft tissue or acoustic-shadow (behind bone surface) region, both learned through image descriptors. Pair-wise edge potentials encode ultrasound propagation constraints of bone surfaces given their large acoustic-impedance difference. We evaluate the proposed method in comparison with four earlier approaches, on in-vivo ultrasound images collected from dorsal and volar views of the forearm. The proposed method achieves an average root-mean-square error and symmetric Hausdorff distance of 0.28mm and 1.78mm, respectively. It detects 99.9% of the annotated bone surfaces with a mean scanline error (distance to annotations) of 0.39mm.
△ Less
Submitted 7 January, 2020;
originally announced January 2020.
-
Active Learning for Segmentation Based on Bayesian Sample Queries
Authors:
Firat Ozdemir,
Zixuan Peng,
Philipp Fuernstahl,
Christine Tanner,
Orcun Goksel
Abstract:
Segmentation of anatomical structures is a fundamental image analysis task for many applications in the medical field. Deep learning methods have been shown to perform well, but for this purpose large numbers of manual annotations are needed in the first place, which necessitate prohibitive levels of resources that are often unavailable. In an active learning framework of selecting informed sample…
▽ More
Segmentation of anatomical structures is a fundamental image analysis task for many applications in the medical field. Deep learning methods have been shown to perform well, but for this purpose large numbers of manual annotations are needed in the first place, which necessitate prohibitive levels of resources that are often unavailable. In an active learning framework of selecting informed samples for manual labeling, expert clinician time for manual annotation can be optimally utilized, enabling the establishment of large labeled datasets for machine learning. In this paper, we propose a novel method that combines representativeness with uncertainty in order to estimate ideal samples to be annotated, iteratively from a given dataset. Our novel representativeness metric is based on Bayesian sampling, by using information-maximizing autoencoders. We conduct experiments on a shoulder magnetic resonance imaging (MRI) dataset for the segmentation of four musculoskeletal tissue classes. Quantitative results show that the annotation of representative samples selected by our proposed querying method yields an improved segmentation performance at each active learning iteration, compared to a baseline method that also employs uncertainty and representativeness metrics. For instance, with only 10% of the dataset annotated, our method reaches within 5% of Dice score expected from the upper bound scenario of all the dataset given as annotated (an impractical scenario due to resource constraints), and this gap drops down to a mere 2% when less than a fifth of the dataset samples are annotated. Such active learning approach to selecting samples to annotate enables an optimal use of the expert clinician time, being often the bottleneck in realizing machine learning solutions in medicine.
△ Less
Submitted 22 December, 2019;
originally announced December 2019.
-
Planet Hunters TESS I: TOI 813, a subgiant hosting a transiting Saturn-sized planet on an 84-day orbit
Authors:
N. L. Eisner,
O. Barragán,
S. Aigrain,
C. Lintott,
G. Miller,
N. Zicher,
T. S. Boyajian,
C. Briceño,
E. M. Bryant,
J. L. Christiansen,
A. D. Feinstein,
L. M. Flor-Torres,
M. Fridlund,
D. Gandolfi,
J. Gilbert,
N. Guerrero,
J. M. Jenkins,
K. Jones,
M. H. Kristiansen,
A. Vanderburg,
N. Law,
A. R. López-Sánchez,
A. W. Mann,
E. J. Safron,
M. E. Schwamb
, et al. (25 additional authors not shown)
Abstract:
We report on the discovery and validation of TOI 813b (TIC 55525572 b), a transiting exoplanet identified by citizen scientists in data from NASA's Transiting Exoplanet Survey Satellite (TESS) and the first planet discovered by the Planet Hunters TESS project. The host star is a bright (V = 10.3 mag) subgiant ($R_\star=1.94\,R_\odot$, $M_\star=1.32\,M_\odot$). It was observed almost continuously b…
▽ More
We report on the discovery and validation of TOI 813b (TIC 55525572 b), a transiting exoplanet identified by citizen scientists in data from NASA's Transiting Exoplanet Survey Satellite (TESS) and the first planet discovered by the Planet Hunters TESS project. The host star is a bright (V = 10.3 mag) subgiant ($R_\star=1.94\,R_\odot$, $M_\star=1.32\,M_\odot$). It was observed almost continuously by TESS during its first year of operations, during which time four individual transit events were detected. The candidate passed all the standard light curve-based vetting checks, and ground-based follow-up spectroscopy and speckle imaging enabled us to place an upper limit of $2 M_{Jup}$ (99 % confidence) on the mass of the companion, and to statistically validate its planetary nature. Detailed modelling of the transits yields a period of $83.8911_{ - 0.0031 } ^ { + 0.0027 }$ days, a planet radius of $6.71 \pm 0.38$ $R_{\oplus}$, and a semi major axis of $0.423_{ - 0.037 } ^ { + 0.031 }$ AU. The planet's orbital period combined with the evolved nature of the host star places this object in a relatively under-explored region of parameter space. We estimate that TOI-813b induces a reflex motion in its host star with a semi-amplitude of $\sim6$ ms$^{-1}$, making this system a promising target to measure the mass of a relatively long-period transiting planet.
△ Less
Submitted 13 January, 2020; v1 submitted 19 September, 2019;
originally announced September 2019.
-
Determination of the scalar and vector polarizabilities of the cesium $6s \ ^2S_{1/2} \rightarrow 7s \ ^2S_{1/2}$ transition and implications for atomic parity non-conservation
Authors:
George Toh,
Amy Damitz,
Carol E. Tanner,
W. R. Johnson,
D. S. Elliott
Abstract:
Using recent high-precision measurements of electric dipole matrix elements of atomic cesium, we make an improved determination of the scalar ($α$) and vector ($β$) polarizabilities of the cesium $6s \ ^2S_{1/2} \rightarrow 7s \ ^2S_{1/2} $ transition calculated through a sum-over-states method. We report values of $α= -268.82 \ (30) \ a_0^3$ and $β= 27.139 \ (42) \ a_0^3$ with the highest precisi…
▽ More
Using recent high-precision measurements of electric dipole matrix elements of atomic cesium, we make an improved determination of the scalar ($α$) and vector ($β$) polarizabilities of the cesium $6s \ ^2S_{1/2} \rightarrow 7s \ ^2S_{1/2} $ transition calculated through a sum-over-states method. We report values of $α= -268.82 \ (30) \ a_0^3$ and $β= 27.139 \ (42) \ a_0^3$ with the highest precision to date. We find a discrepancy between our value of $β$ and the past preferred value, resulting in a significant shift in the value of the weak charge $Q_w$ of the cesium nucleus. Future work to resolve the differences in the polarizability will be critical for interpretation of parity non-conservation measurements in cesium, which have implications for physics beyond the Standard Model.
△ Less
Submitted 25 May, 2019; v1 submitted 7 May, 2019;
originally announced May 2019.
-
Measurement of the radial matrix elements for the $6s ^2S_{1/2} \rightarrow 7p ^2P_J$ transitions in cesium
Authors:
Amy Damitz,
George Toh,
Eric Putney,
Carol E. Tanner,
D. S. Elliott
Abstract:
We report measurements of the electric dipole matrix elements of the $^{133}$Cs $\ $ $6s\,^2S_{1/2} \rightarrow 7p\,^2P_{1/2}$ and $6s\,^2S_{1/2} \rightarrow 7p\,^2P_{3/2}$ transitions. Each of these determinations is based on direct, precise comparisons of the absorption coefficients between two absorption lines. For the $\langle 6s\,^2S_{1/2}||r|| 7p\,^2P_{3/2} \rangle$ matrix element, we measur…
▽ More
We report measurements of the electric dipole matrix elements of the $^{133}$Cs $\ $ $6s\,^2S_{1/2} \rightarrow 7p\,^2P_{1/2}$ and $6s\,^2S_{1/2} \rightarrow 7p\,^2P_{3/2}$ transitions. Each of these determinations is based on direct, precise comparisons of the absorption coefficients between two absorption lines. For the $\langle 6s\,^2S_{1/2}||r|| 7p\,^2P_{3/2} \rangle$ matrix element, we measure the ratio of the absorption coefficient on this line with that of the D$_1$ transition, $6s\,^2S_{1/2} \rightarrow 6p\,^2P_{1/2}$. The matrix element of the D$_1$ line has been determined with high precision previously by many groups. For the $\langle 6s\,^2S_{1/2}||r|| 7p\,^2P_{1/2} \rangle$ matrix element, we measure the ratio of the absorption coefficient on this line with that of the $6s\,^2S_{1/2} \rightarrow 7p\,^2P_{3/2}$ transition. Our results for these matrix elements are $\langle 6s\,^2S_{1/2}||r|| 7p\,^2P_{3/2} \rangle = 0.57417 \: (57)~a_0$ and $\langle 6s\,^2S_{1/2}||r|| 7p\,^2P_{1/2} \rangle = 0.27810 \: (45)~a_0$. These measurements have implications for the interpretation of parity nonconservation in atoms.
△ Less
Submitted 12 April, 2019;
originally announced April 2019.
-
Catalog of New K2 Exoplanet Candidates from Citizen Scientists
Authors:
Jon K. Zink,
Kevin K. Hardegree-Ullman,
Jessie L. Christiansen,
Ian J. M. Crossfield,
Erik A. Petigura,
Chris J. Lintott,
John H. Livingston,
David R. Ciardi,
Geert Barentsen,
Courtney D. Dressing,
Alexander Ye,
Joshua E. Schlieder,
Kevin Acres,
Peter Ansorge,
Dario Arienti,
Elisabeth Baeten,
Victoriano Canales Cerd,
Itayi Chitsiga,
Maxwell Daly,
James Damboiu,
Martin Ende,
Adnan Erdag,
Stiliyan Evstatiev,
Joseph Henderson,
David Hine
, et al. (12 additional authors not shown)
Abstract:
We provide 28 new planet candidates that have been vetted by citizen scientists and expert astronomers. This catalog contains 9 likely rocky candidates ($R_{pl} < 2.0R_\oplus$) and 19 gaseous candidates ($R_{pl} > 2.0R_\oplus$). Within this list we find one multi-planet system (EPIC 246042088). These two sub-Neptune ($2.99 \pm 0.02R_\oplus$ and $3.44 \pm 0.02R_\oplus$) planets exist in a near 3:2…
▽ More
We provide 28 new planet candidates that have been vetted by citizen scientists and expert astronomers. This catalog contains 9 likely rocky candidates ($R_{pl} < 2.0R_\oplus$) and 19 gaseous candidates ($R_{pl} > 2.0R_\oplus$). Within this list we find one multi-planet system (EPIC 246042088). These two sub-Neptune ($2.99 \pm 0.02R_\oplus$ and $3.44 \pm 0.02R_\oplus$) planets exist in a near 3:2 orbital resonance. The discovery of this multi-planet system is important in its addition to the list of known multi-planet systems within the K2 catalog, and more broadly in understanding the multiplicity distribution of the exoplanet population (Zink et al. 2019). The candidates on this list are anticipated to generate RV amplitudes of 0.2-18 m/s, many within the range accessible to current facilities.
△ Less
Submitted 5 March, 2019; v1 submitted 1 March, 2019;
originally announced March 2019.
-
K2-288Bb: A Small Temperate Planet in a Low-mass Binary System Discovered by Citizen Scientists
Authors:
Adina D. Feinstein,
Joshua E. Schlieder,
John H. Livingston,
David R. Ciardi,
Andrew W. Howard,
Lauren Arnold,
Geert Barentsen,
Makennah Bristow,
Jessie L. Christiansen,
Ian J. M. Crossfield,
Courtney D. Dressing,
Erica J. Gonzales,
Molly Kosiarek,
Chris J. Lintott,
Grant Miller,
Farisa Y. Morales,
Erik A. Petigura,
Beverly Thackeray,
Joanne Ault,
Elisabeth Baeten,
Alexander F. Jonkeren,
James Langley,
Houssen Moshinaly,
Kirk Pearson,
Christopher Tanner
, et al. (1 additional authors not shown)
Abstract:
Observations from the Kepler and K2 missions have provided the astronomical community with unprecedented amounts of data to search for transiting exoplanets and other astrophysical phenomena. Here, we present K2-288, a low-mass binary system (M2.0 +/- 1.0; M3.0 +/- 1.0) hosting a small (Rp = 1.9 REarth), temperate (Teq = 226 K) planet observed in K2 Campaign 4. The candidate was first identified b…
▽ More
Observations from the Kepler and K2 missions have provided the astronomical community with unprecedented amounts of data to search for transiting exoplanets and other astrophysical phenomena. Here, we present K2-288, a low-mass binary system (M2.0 +/- 1.0; M3.0 +/- 1.0) hosting a small (Rp = 1.9 REarth), temperate (Teq = 226 K) planet observed in K2 Campaign 4. The candidate was first identified by citizen scientists using Exoplanet Explorers hosted on the Zooniverse platform. Follow-up observations and detailed analyses validate the planet and indicate that it likely orbits the secondary star on a 31.39-day period. This orbit places K2-288Bb in or near the habitable zone of its low-mass host star. K2-288Bb resides in a system with a unique architecture, as it orbits at >0.1 au from one component in a moderate separation binary (aproj approximately 55 au), and further follow-up may provide insight into its formation and evolution. Additionally, its estimated size straddles the observed gap in the planet radius distribution. Planets of this size occur less frequently and may be in a transient phase of radius evolution. K2-288 is the third transiting planet system identified by the Exoplanet Explorers program and its discovery exemplifies the value of citizen science in the era of Kepler, K2, and the Transiting Exoplanet Survey Satellite.
△ Less
Submitted 7 February, 2019;
originally announced February 2019.
-
SCATGAN for Reconstruction of Ultrasound Scatterers Using Generative Adversarial Networks
Authors:
Andrawes Al Bahou,
Christine Tanner,
Orcun Goksel
Abstract:
Computational simulation of ultrasound (US) echography is essential for training sonographers. Realistic simulation of US interaction with microscopic tissue structures is often modeled by a tissue representation in the form of point scatterers, convolved with a spatially varying point spread function. This yields a realistic US B-mode speckle texture, given that a scatterer representation for a p…
▽ More
Computational simulation of ultrasound (US) echography is essential for training sonographers. Realistic simulation of US interaction with microscopic tissue structures is often modeled by a tissue representation in the form of point scatterers, convolved with a spatially varying point spread function. This yields a realistic US B-mode speckle texture, given that a scatterer representation for a particular tissue type is readily available. This is often not the case and scatterers are nontrivial to determine. In this work we propose to estimate scatterer maps from sample US B-mode images of a tissue, by formulating this inverse mapping problem as image translation, where we learn the mapping with Generative Adversarial Networks, using a US simulation software for training. We demonstrate robust reconstruction results, invariant to US viewing and imaging settings such as imaging direction and center frequency. Our method is shown to generalize beyond the trained imaging settings, demonstrated on in-vivo US data. Our inference runs orders of magnitude faster than optimization-based techniques, enabling future extensions for reconstructing 3D B-mode volumes with only linear computational complexity.
△ Less
Submitted 1 February, 2019;
originally announced February 2019.
-
Siamese Networks with Location Prior for Landmark Tracking in Liver Ultrasound Sequences
Authors:
Alvaro Gomariz,
Weiye Li,
Ece Ozkan,
Christine Tanner,
Orcun Goksel
Abstract:
Image-guided radiation therapy can benefit from accurate motion tracking by ultrasound imaging, in order to minimize treatment margins and radiate moving anatomical targets, e.g., due to breathing. One way to formulate this tracking problem is the automatic localization of given tracked anatomical landmarks throughout a temporal ultrasound sequence. For this, we herein propose a fully-convolutiona…
▽ More
Image-guided radiation therapy can benefit from accurate motion tracking by ultrasound imaging, in order to minimize treatment margins and radiate moving anatomical targets, e.g., due to breathing. One way to formulate this tracking problem is the automatic localization of given tracked anatomical landmarks throughout a temporal ultrasound sequence. For this, we herein propose a fully-convolutional Siamese network that learns the similarity between pairs of image regions containing the same landmark. Accordingly, it learns to localize and thus track arbitrary image features, not only predefined anatomical structures. We employ a temporal consistency model as a location prior, which we combine with the network-predicted location probability map to track a target iteratively in ultrasound sequences. We applied this method on the dataset of the Challenge on Liver Ultrasound Tracking (CLUST) with competitive results, where our work is the first to effectively apply CNNs on this tracking problem, thanks to our temporal regularization.
△ Less
Submitted 23 January, 2019;
originally announced January 2019.
-
Iterative Interaction Training for Segmentation Editing Networks
Authors:
Gustav Bredell,
Christine Tanner,
Ender Konukoglu
Abstract:
Automatic segmentation has great potential to facilitate morphological measurements while simultaneously increasing efficiency. Nevertheless often users want to edit the segmentation to their own needs and will need different tools for this. There has been methods developed to edit segmentations of automatic methods based on the user input, primarily for binary segmentations. Here however, we pres…
▽ More
Automatic segmentation has great potential to facilitate morphological measurements while simultaneously increasing efficiency. Nevertheless often users want to edit the segmentation to their own needs and will need different tools for this. There has been methods developed to edit segmentations of automatic methods based on the user input, primarily for binary segmentations. Here however, we present an unique training strategy for convolutional neural networks (CNNs) trained on top of an automatic method to enable interactive segmentation editing that is not limited to binary segmentation. By utilizing a robot-user during training, we closely mimic realistic use cases to achieve optimal editing performance. In addition, we show that an increase of the iterative interactions during the training process up to ten improves the segmentation editing performance substantially. Furthermore, we compare our segmentation editing CNN (interCNN) to state-of-the-art interactive segmentation algorithms and show a superior or on par performance.
△ Less
Submitted 23 July, 2018;
originally announced July 2018.
-
Generative Adversarial Networks for MR-CT Deformable Image Registration
Authors:
Christine Tanner,
Firat Ozdemir,
Romy Profanter,
Valeriy Vishnevsky,
Ender Konukoglu,
Orcun Goksel
Abstract:
Deformable Image Registration (DIR) of MR and CT images is one of the most challenging registration task, due to the inherent structural differences of the modalities and the missing dense ground truth. Recently cycle Generative Adversarial Networks (cycle-GANs) have been used to learn the intensity relationship between these 2 modalities for unpaired brain data. Yet its usefulness for DIR was not…
▽ More
Deformable Image Registration (DIR) of MR and CT images is one of the most challenging registration task, due to the inherent structural differences of the modalities and the missing dense ground truth. Recently cycle Generative Adversarial Networks (cycle-GANs) have been used to learn the intensity relationship between these 2 modalities for unpaired brain data. Yet its usefulness for DIR was not assessed.
In this study we evaluate the DIR performance for thoracic and abdominal organs after synthesis by cycle-GAN. We show that geometric changes, which differentiate the two populations (e.g. inhale vs. exhale), are readily synthesized as well. This causes substantial problems for any application which relies on spatial correspondences being preserved between the real and the synthesized image (e.g. plan, segmentation, landmark propagation). To alleviate this problem, we investigated reducing the spatial information provided to the discriminator by decreasing the size of its receptive fields.
Image synthesis was learned from 17 unpaired subjects per modality. Registration performance was evaluated with respect to manual segmentations of 11 structures for 3 subjects from the VISERAL challenge. State-of-the-art DIR methods based on Normalized Mutual Information (NMI), Modality Independent Neighborhood Descriptor (MIND) and their novel combination achieved a mean segmentation overlap ratio of 76.7, 67.7, 76.9%, respectively. This dropped to 69.1% or less when registering images synthesized by cycle-GAN based on local correlation, due to the poor performance on the thoracic region, where large lung volume changes were synthesized. Performance for the abdominal region was similar to that of CT-MRI NMI registration (77.4 vs. 78.8%) when using 3D synthesizing MRIs (12 slices) and medium sized receptive fields for the discriminator.
△ Less
Submitted 19 July, 2018;
originally announced July 2018.
-
Active Learning for Segmentation by Optimizing Content Information for Maximal Entropy
Authors:
Firat Ozdemir,
Zixuan Peng,
Christine Tanner,
Philipp Fuernstahl,
Orcun Goksel
Abstract:
Segmentation is essential for medical image analysis tasks such as intervention planning, therapy guidance, diagnosis, treatment decisions. Deep learning is becoming increasingly prominent for segmentation, where the lack of annotations, however, often becomes the main limitation. Due to privacy concerns and ethical considerations, most medical datasets are created, curated, and allow access only…
▽ More
Segmentation is essential for medical image analysis tasks such as intervention planning, therapy guidance, diagnosis, treatment decisions. Deep learning is becoming increasingly prominent for segmentation, where the lack of annotations, however, often becomes the main limitation. Due to privacy concerns and ethical considerations, most medical datasets are created, curated, and allow access only locally. Furthermore, current deep learning methods are often suboptimal in translating anatomical knowledge between different medical imaging modalities. Active learning can be used to select an informed set of image samples to request for manual annotation, in order to best utilize the limited annotation time of clinical experts for optimal outcomes, which we focus on in this work. Our contributions herein are two fold: (1) we enforce domain-representativeness of selected samples using a proposed penalization scheme to maximize information at the network abstraction layer, and (2) we propose a Borda-count based sample querying scheme for selecting samples for segmentation. Comparative experiments with baseline approaches show that the samples queried with our proposed method, where both above contributions are combined, result in significantly improved segmentation performance for this active learning task.
△ Less
Submitted 18 July, 2018;
originally announced July 2018.
-
Temporal Interpolation via Motion Field Prediction
Authors:
Lin Zhang,
Neerav Karani,
Christine Tanner,
Ender Konukoglu
Abstract:
Navigated 2D multi-slice dynamic Magnetic Resonance (MR) imaging enables high contrast 4D MR imaging during free breathing and provides in-vivo observations for treatment planning and guidance. Navigator slices are vital for retrospective stacking of 2D data slices in this method. However, they also prolong the acquisition sessions. Temporal interpolation of navigator slices an be used to reduce t…
▽ More
Navigated 2D multi-slice dynamic Magnetic Resonance (MR) imaging enables high contrast 4D MR imaging during free breathing and provides in-vivo observations for treatment planning and guidance. Navigator slices are vital for retrospective stacking of 2D data slices in this method. However, they also prolong the acquisition sessions. Temporal interpolation of navigator slices an be used to reduce the number of navigator acquisitions without degrading specificity in stacking. In this work, we propose a convolutional neural network (CNN) based method for temporal interpolation via motion field prediction. The proposed formulation incorporates the prior knowledge that a motion field underlies changes in the image intensities over time. Previous approaches that interpolate directly in the intensity space are prone to produce blurry images or even remove structures in the images. Our method avoids such problems and faithfully preserves the information in the image. Further, an important advantage of our formulation is that it provides an unsupervised estimation of bi-directional motion fields. We show that these motion fields can be used to halve the number of registrations required during 4D reconstruction, thus substantially reducing the reconstruction time.
△ Less
Submitted 12 April, 2018;
originally announced April 2018.