-
Cavity-assisted magnetization switching in a quantum spin-phonon chain
Authors:
Mohsen Yarmohammadi,
Peter M. Oppeneer,
James K. Freericks
Abstract:
We propose a Néel magnetization switching mechanism in a hybrid magnon-phonon optical cavity system. A terahertz-pumped single-mode cavity photon couples to a spin-phonon chain, while the system dissipates energy via different baths. Our mean-field analysis reveals that the photon induces magnetization switching by generating strongly entangled magnon pairs with opposite momenta -- a feature weakl…
▽ More
We propose a Néel magnetization switching mechanism in a hybrid magnon-phonon optical cavity system. A terahertz-pumped single-mode cavity photon couples to a spin-phonon chain, while the system dissipates energy via different baths. Our mean-field analysis reveals that the photon induces magnetization switching by generating strongly entangled magnon pairs with opposite momenta -- a feature weakly present in the cavity-free system. This switching occurs only at specific drive frequencies, namely at low magnon energies and near the upper edge (perpendicular modes) of the two-magnon band. Our results underscore the roles of laser fluence, damping, and photon loss in modulating the switching process, offering a promising route for cavity-assisted magnetization control in opto-spintronics.
△ Less
Submitted 28 March, 2025;
originally announced March 2025.
-
Anisotropic light-tailored RKKY interaction in two-dimensional $d$-wave altermagnets
Authors:
Mohsen Yarmohammadi,
Ulrich Zülicke,
Jamal Berakdar,
Jacob Linder,
James K. Freericks
Abstract:
Altermagnets are known in spintronics for their intrinsic spin-splitting and unconventional magnetic responses, particularly to magnetic impurities. However, effectively controlling the magnetic exchange interactions in altermagnets is challenging for practical applications. Here, we propose using circularly polarized light to tune the Ruderman-Kittel-Kasuya-Yosida (RKKY) interaction in two-dimens…
▽ More
Altermagnets are known in spintronics for their intrinsic spin-splitting and unconventional magnetic responses, particularly to magnetic impurities. However, effectively controlling the magnetic exchange interactions in altermagnets is challenging for practical applications. Here, we propose using circularly polarized light to tune the Ruderman-Kittel-Kasuya-Yosida (RKKY) interaction in two-dimensional $d$-wave altermagnets. Using the real-space retarded Green's functions approach, our results show that while the Heisenberg and Ising exchanges dominate, a notable Dzyaloshinskii-Moriya (DM) interaction also plays a key role. Furthermore, the inherent strength of altermagnetism imprints chirp-like signatures into the magnetic responses, which can be dynamically tuned via light. We mainly demonstrate that gate-induced Rashba spin-orbit coupling is essential in response to light -- light selectively and anisotropically adjusts the DM interaction without affecting the other exchanges. Our findings further indicate that rotating the altermagnet by $45^\circ$ relative to the light's polarization direction generates a Dirac-like dispersion and different DM interactions. We finally extract critical thresholds where light reverses DM interactions along one axis or balances both in-plane components. The anisotropic light-driven control of RKKY interactions in 2D altermagnets not only highlights their unique properties but also opens new avenues for engineering tailored magnetic characteristics in spintronic applications.
△ Less
Submitted 27 March, 2025;
originally announced March 2025.
-
mFollowIR: a Multilingual Benchmark for Instruction Following in Retrieval
Authors:
Orion Weller,
Benjamin Chang,
Eugene Yang,
Mahsa Yarmohammadi,
Sam Barham,
Sean MacAvaney,
Arman Cohan,
Luca Soldaini,
Benjamin Van Durme,
Dawn Lawrie
Abstract:
Retrieval systems generally focus on web-style queries that are short and underspecified. However, advances in language models have facilitated the nascent rise of retrieval models that can understand more complex queries with diverse intents. However, these efforts have focused exclusively on English; therefore, we do not yet understand how they work across languages. We introduce mFollowIR, a mu…
▽ More
Retrieval systems generally focus on web-style queries that are short and underspecified. However, advances in language models have facilitated the nascent rise of retrieval models that can understand more complex queries with diverse intents. However, these efforts have focused exclusively on English; therefore, we do not yet understand how they work across languages. We introduce mFollowIR, a multilingual benchmark for measuring instruction-following ability in retrieval models. mFollowIR builds upon the TREC NeuCLIR narratives (or instructions) that span three diverse languages (Russian, Chinese, Persian) giving both query and instruction to the retrieval models. We make small changes to the narratives and isolate how well retrieval models can follow these nuanced changes. We present results for both multilingual (XX-XX) and cross-lingual (En-XX) performance. We see strong cross-lingual performance with English-based retrievers that trained using instructions, but find a notable drop in performance in the multilingual setting, indicating that more work is needed in developing data for instruction-based multilingual retrievers.
△ Less
Submitted 31 January, 2025;
originally announced January 2025.
-
Topological anomalous Floquet photon pump
Authors:
Manshuo Lin,
Saeed Rahmanian Koshkaki,
Mohsen Yarmohammadi,
Michael H. Kolodrubetz
Abstract:
There is a close theoretical connection between topological Floquet physics and cavity QED, yet this connection has not been realized experimentally due to complicated cavity QED models that often arise. We propose a simple, experimentally viable protocol to realize non-adiabatic topological photon pumping mediated by a single qubit, which we dub the anomalous Floquet photon pump. For both quantiz…
▽ More
There is a close theoretical connection between topological Floquet physics and cavity QED, yet this connection has not been realized experimentally due to complicated cavity QED models that often arise. We propose a simple, experimentally viable protocol to realize non-adiabatic topological photon pumping mediated by a single qubit, which we dub the anomalous Floquet photon pump. For both quantized photons and external drive, the system exhibits a non-trivial topological phase across a broad range of parameter space. Transitions out of the topological phase result from frequency-space delocalization. Finally, we argue that the protocol can be implemented in existing experiments via driven qubit non-linearities, with topological pumping witnessed in measurements of the cavity Wigner distribution functions.
△ Less
Submitted 7 November, 2024; v1 submitted 14 October, 2024;
originally announced October 2024.
-
Probing topological phases in a perturbed Kane-Mele model via RKKY interaction: Application to monolayer jacutingaite Pt$_2$HgSe$_3$
Authors:
Mohsen Yarmohammadi,
Saeed Rahmanian Koshkaki,
Jamal Berakdar,
Marin Bukov,
Michael H. Kolodrubetz
Abstract:
Quantum spin Hall insulators (QSHIs) are promising for spintronics, leveraging strong spin-orbit coupling for efficient spin manipulation via electrical and optical methods, with potential applications in memory storage, quantum computing, and spin-based logic. While the Kane-Mele model effectively captures the low-energy physics of these materials, the impact of perturbations driving phase transi…
▽ More
Quantum spin Hall insulators (QSHIs) are promising for spintronics, leveraging strong spin-orbit coupling for efficient spin manipulation via electrical and optical methods, with potential applications in memory storage, quantum computing, and spin-based logic. While the Kane-Mele model effectively captures the low-energy physics of these materials, the impact of perturbations driving phase transitions is less well understood. Developing approaches to describe these phases is crucial. Here, we study the noncollinear Ruderman-Kittel-Kasuya-Yosida (RKKY) interaction between two magnetic impurities in a \textit{perturbed} Kane-Mele model with strong spin-orbit coupling, relevant to monolayer jacutingaite Pt$_2$HgSe$_3$ as a prominent QSHI. By analyzing RKKY interactions, we reveal distinct, relative (rather than absolute) signatures of the different phase transitions induced by static and dynamic perturbations based on their impact on magnetic impurities. We then employ these perturbations to switch between ferromagnetic and antiferromagnetic or clockwise and counterclockwise magnetic interactions for control processes. Our results offer a practical way to track topological phases via magnetic properties.
△ Less
Submitted 15 October, 2024; v1 submitted 14 October, 2024;
originally announced October 2024.
-
Terahertz high-harmonic generation in gapped antiferromagnetic chains
Authors:
Mohsen Yarmohammadi,
Michael H. Kolodrubetz
Abstract:
The nonlinear dynamics of magnetization in antiferromagnets, resulting in high-frequency spin waves (high-order harmonics) as signal carriers, enable fast magnetic state switching in spintronic devices. More harmonic orders potentially allow more information to be conveyed by the spins. Developing theoretical models to describe these waves in antiferromagnets is essential for predicting their prop…
▽ More
The nonlinear dynamics of magnetization in antiferromagnets, resulting in high-frequency spin waves (high-order harmonics) as signal carriers, enable fast magnetic state switching in spintronic devices. More harmonic orders potentially allow more information to be conveyed by the spins. Developing theoretical models to describe these waves in antiferromagnets is essential for predicting their properties and guiding experimental efforts. Here, we consider the role of linear and quadratic spin-phonon couplings (SPCs) in achieving high-order harmonics in the THz magnetization of a gapped antiferromagnetic spin chain. A THz steady laser's electric field indirectly drives spins via phonons. Using spin-wave theory, mean-field theory, and the Lindblad formalism, we analyze the resulting nonlinear dynamics. We highlight the distinct mechanisms for harmonic generation when a phonon is coupled to the easy-plane and easy-axis of spins. Moreover, we observe that quadratic SPC blocks odd harmonics due to invariant inversion symmetry, while linear SPC generates both odd and even harmonics. We also investigate the effects of drive frequency, drive amplitude, phonon damping, and spin damping on the number of harmonics. Our findings offer an alternative pathway for developing nonlinear magnonics.
△ Less
Submitted 2 August, 2024;
originally announced August 2024.
-
Improving Neural Biasing for Contextual Speech Recognition by Early Context Injection and Text Perturbation
Authors:
Ruizhe Huang,
Mahsa Yarmohammadi,
Sanjeev Khudanpur,
Daniel Povey
Abstract:
Existing research suggests that automatic speech recognition (ASR) models can benefit from additional contexts (e.g., contact lists, user specified vocabulary). Rare words and named entities can be better recognized with contexts. In this work, we propose two simple yet effective techniques to improve context-aware ASR models. First, we inject contexts into the encoders at an early stage instead o…
▽ More
Existing research suggests that automatic speech recognition (ASR) models can benefit from additional contexts (e.g., contact lists, user specified vocabulary). Rare words and named entities can be better recognized with contexts. In this work, we propose two simple yet effective techniques to improve context-aware ASR models. First, we inject contexts into the encoders at an early stage instead of merely at their last layers. Second, to enforce the model to leverage the contexts during training, we perturb the reference transcription with alternative spellings so that the model learns to rely on the contexts to make correct predictions. On LibriSpeech, our techniques together reduce the rare word error rate by 60% and 25% relatively compared to no biasing and shallow fusion, making the new state-of-the-art performance. On SPGISpeech and a real-world dataset ConEC, our techniques also yield good improvements over the baselines.
△ Less
Submitted 14 July, 2024;
originally announced July 2024.
-
Spin high-harmonic generation through terahertz laser-driven phonons
Authors:
Negin Moharrami Allafi,
Michael H. Kolodrubetz,
Marin Bukov,
Vadim Oganesyan,
Mohsen Yarmohammadi
Abstract:
In the realm of open quantum systems, steady states and high-harmonic generation (HHG) existing far from equilibrium have become core pillars of ultrafast science. Most solid-state research explores charge HHG with limited investigations into spin degrees of freedom. In this study, we theoretically address spin HHG in the steady state resulting from the terahertz laser-driven spin-phonon coupling…
▽ More
In the realm of open quantum systems, steady states and high-harmonic generation (HHG) existing far from equilibrium have become core pillars of ultrafast science. Most solid-state research explores charge HHG with limited investigations into spin degrees of freedom. In this study, we theoretically address spin HHG in the steady state resulting from the terahertz laser-driven spin-phonon coupling in a dissipative dimerized spin-1/2 chain. Instead of directly driving spins using time-dependent magnetic fields, we employ the magnetophononic mechanism, where the laser first drives the lattice, and then the excited lattice subsequently drives the spins. We investigate the role of various model parameters for optimizing HHG. Increasing the laser's amplitude amplifies spin HHG beyond the perturbative regime, enhancing both harmonic amplitudes and orders. We find that configuring the drive frequency far below the spin band yields the highest harmonic order. Additionally, we provide a theory matching the numerical results under weak spin-phonon coupling and propose an experimental procedure to probe the emission spectrum of spin HHG.
△ Less
Submitted 8 April, 2024;
originally announced April 2024.
-
Ultrafast dynamics of a fermion chain in a terahertz field-driven optical cavity
Authors:
Mohsen Yarmohammadi,
John Sous,
Marin Bukov,
Michael H. Kolodrubetz
Abstract:
We study the effect of a terahertz field-driven single cavity mode for ultrafast control of a fermion chain with dissipation-induced nonlinearity and quadratic coupling to an infrared-active phonon mode. Without photon loss from the cavity, we uncover a first-order phase transition in the nonequilibrium steady state only for the lower phonon-polariton, accompanied by polaritons whose frequency res…
▽ More
We study the effect of a terahertz field-driven single cavity mode for ultrafast control of a fermion chain with dissipation-induced nonlinearity and quadratic coupling to an infrared-active phonon mode. Without photon loss from the cavity, we uncover a first-order phase transition in the nonequilibrium steady state only for the lower phonon-polariton, accompanied by polaritons whose frequency response is asymmetric with respect to the photon frequency due to the direct laser-induced dressing effect on the photon. A weak laser field fails to induce the phase transition but renders the polaritons symmetrical. Finally, we show that sufficiently strong photon loss from the cavity eliminates the polaritons and the associated phase transition. The experimental feasibility of these phenomena is also proposed.
△ Less
Submitted 19 February, 2024;
originally announced February 2024.
-
MultiMUC: Multilingual Template Filling on MUC-4
Authors:
William Gantt,
Shabnam Behzad,
Hannah YoungEun An,
Yunmo Chen,
Aaron Steven White,
Benjamin Van Durme,
Mahsa Yarmohammadi
Abstract:
We introduce MultiMUC, the first multilingual parallel corpus for template filling, comprising translations of the classic MUC-4 template filling benchmark into five languages: Arabic, Chinese, Farsi, Korean, and Russian. We obtain automatic translations from a strong multilingual machine translation system and manually project the original English annotations into each target language. For all la…
▽ More
We introduce MultiMUC, the first multilingual parallel corpus for template filling, comprising translations of the classic MUC-4 template filling benchmark into five languages: Arabic, Chinese, Farsi, Korean, and Russian. We obtain automatic translations from a strong multilingual machine translation system and manually project the original English annotations into each target language. For all languages, we also provide human translations for sentences in the dev and test splits that contain annotated template arguments. Finally, we present baselines on MultiMUC both with state-of-the-art template filling models and with ChatGPT.
△ Less
Submitted 29 January, 2024;
originally announced January 2024.
-
Laser-enhanced magnetism in SmFeO$_3$
Authors:
Mohsen Yarmohammadi,
Marin Bukov,
Vadim Oganesyan,
Michael H. Kolodrubetz
Abstract:
To coherently enhance inherent weak magnetic interactions in rare-earth orthoferrite SmFeO$_3$ as a functional material for spintronic applications, we simulate the dissipative spin dynamics that are linearly and quadratically coupled to laser-driven infrared-active phonons. When linear coupling dominates, we discover a magnetophononic dynamical first-order phase transition in the nonequilibrium s…
▽ More
To coherently enhance inherent weak magnetic interactions in rare-earth orthoferrite SmFeO$_3$ as a functional material for spintronic applications, we simulate the dissipative spin dynamics that are linearly and quadratically coupled to laser-driven infrared-active phonons. When linear coupling dominates, we discover a magnetophononic dynamical first-order phase transition in the nonequilibrium steady state which can inhibit strong enhancement of magnetic interactions. By contrast, when quadratic spin-phonon coupling dominates, no phase transition exists at experimentally relevant parameters. By utilizing a chirp protocol, the phase transition can be engineered, enabling stronger magnetic interactions. We also discuss the route for experimental observation of our results.
△ Less
Submitted 1 October, 2023; v1 submitted 28 August, 2023;
originally announced August 2023.
-
MegaWika: Millions of reports and their sources across 50 diverse languages
Authors:
Samuel Barham,
Orion Weller,
Michelle Yuan,
Kenton Murray,
Mahsa Yarmohammadi,
Zhengping Jiang,
Siddharth Vashishtha,
Alexander Martin,
Anqi Liu,
Aaron Steven White,
Jordan Boyd-Graber,
Benjamin Van Durme
Abstract:
To foster the development of new models for collaborative AI-assisted report generation, we introduce MegaWika, consisting of 13 million Wikipedia articles in 50 diverse languages, along with their 71 million referenced source materials. We process this dataset for a myriad of applications, going beyond the initial Wikipedia citation extraction and web scraping of content, including translating no…
▽ More
To foster the development of new models for collaborative AI-assisted report generation, we introduce MegaWika, consisting of 13 million Wikipedia articles in 50 diverse languages, along with their 71 million referenced source materials. We process this dataset for a myriad of applications, going beyond the initial Wikipedia citation extraction and web scraping of content, including translating non-English articles for cross-lingual applications and providing FrameNet parses for automated semantic analysis. MegaWika is the largest resource for sentence-level report generation and the only report generation dataset that is multilingual. We manually analyze the quality of this resource through a semantically stratified sample. Finally, we provide baseline results and trained models for crucial steps in automated report generation: cross-lingual question answering and citation retrieval.
△ Less
Submitted 13 July, 2023;
originally announced July 2023.
-
Nonequilibrium phononic first-order phase transition in a driven fermion chain
Authors:
Mohsen Yarmohammadi,
Marin Bukov,
Michael H. Kolodrubetz
Abstract:
We study the direct laser drive of infrared-active phonons that are quadratically coupled to a spinless fermion chain. Feedback is incorporated by phonon dressing of the electronic dispersion, which enables effective non-linearities in the phonon dynamics. We uncover a first-order phase transition in the phononic steady state in which hysteretic effects allow either large or small phonon occupatio…
▽ More
We study the direct laser drive of infrared-active phonons that are quadratically coupled to a spinless fermion chain. Feedback is incorporated by phonon dressing of the electronic dispersion, which enables effective non-linearities in the phonon dynamics. We uncover a first-order phase transition in the phononic steady state in which hysteretic effects allow either large or small phonon occupation depending on the drive protocol. We discuss the implications of these findings for probing phase transitions in real driven materials.
△ Less
Submitted 29 September, 2023; v1 submitted 25 April, 2023;
originally announced April 2023.
-
Strong-coupling magnetophononics: Self-blocking, phonon-bitriplons, and spin-band engineering
Authors:
M. Yarmohammadi,
M. Krebs,
G. S. Uhrig,
B. Normand
Abstract:
Magnetophononics, the modulation of magnetic interactions by driving infrared-active lattice excitations, is emerging as a key mechanism for the ultrafast dynamical control of both semiclassical and quantum spin systems by coherent light. We demonstrate that, in a quantum magnet with strong spin-phonon coupling, resonances between the driven phonon and the spin excitation frequencies exhibit an in…
▽ More
Magnetophononics, the modulation of magnetic interactions by driving infrared-active lattice excitations, is emerging as a key mechanism for the ultrafast dynamical control of both semiclassical and quantum spin systems by coherent light. We demonstrate that, in a quantum magnet with strong spin-phonon coupling, resonances between the driven phonon and the spin excitation frequencies exhibit an intrinsic self-blocking effect, whereby only a fraction of the available laser power is absorbed by the phonon. Using the quantum master equations governing the nonequilibrium steady states of the coupled spin-lattice system, we show how self-blocking arises from the self-consistent alteration of the resonance frequencies. We link this to the appearance of mutually repelling collective spin-phonon states, which in the regime of strong hybridization become composites of a phonon and two triplons. We then identify the mechanism and optimal phonon frequencies by which to control a global nonequilibrium renormalization of the lattice-driven spin excitation spectrum and demonstrate that this effect should be observable in ultrafast THz experiments on a number of known quantum magnetic materials.
△ Less
Submitted 28 February, 2023;
originally announced March 2023.
-
Truncated Wigner approximation for the bosonic model of large spin baths
Authors:
Mohsen Yarmohammadi,
Katrin Bolsmann,
Yvonne Ribbeheger,
Timo Gräßer,
Götz S. Uhrig
Abstract:
The central spin model has a wide applicability, it is ideally suited to describe a small quantum system, for instance a quantum bit, in contact to a bath of spins, e.g., nuclear spins, or other small quantum systems in general. According to previous work~[Röhrig \textit{et al.}, Phys. Rev. B {\bf 97}, 165431 (2018)], a large bath of quantum spins can be described as a bath of quantum harmonic osc…
▽ More
The central spin model has a wide applicability, it is ideally suited to describe a small quantum system, for instance a quantum bit, in contact to a bath of spins, e.g., nuclear spins, or other small quantum systems in general. According to previous work~[Röhrig \textit{et al.}, Phys. Rev. B {\bf 97}, 165431 (2018)], a large bath of quantum spins can be described as a bath of quantum harmonic oscillators. But the resulting quantum model is still far from being straightforward solvable. Hence we consider a chain representation for the bosonic degrees of freedom to study how well a truncated Wigner approximation of the effective model of harmonic oscillators works in comparison with other approximate and exact methods. Numerically, we examine the effect of the number of bath spins and of the truncation level, i.e., the chain length.
△ Less
Submitted 16 January, 2023;
originally announced January 2023.
-
Noncollinear twisted RKKY interaction on the optically driven SnTe(001) surface
Authors:
Mohsen Yarmohammadi,
Marin Bukov,
Michael H. Kolodrubetz
Abstract:
The nontrivial spin texture on the (001) surface of topological crystalline insulator SnTe hosts exotic scientific importance and spintronic applications. Here, we study the effects of weak Floquet optical driving on the Ruderman-Kittel-Kasuya-Yosida (RKKY) interaction between two magnetic impurities on a doped SnTe(001) surface. Due to peculiar spin-orbit hybridization, we find a noncollinear twi…
▽ More
The nontrivial spin texture on the (001) surface of topological crystalline insulator SnTe hosts exotic scientific importance and spintronic applications. Here, we study the effects of weak Floquet optical driving on the Ruderman-Kittel-Kasuya-Yosida (RKKY) interaction between two magnetic impurities on a doped SnTe(001) surface. Due to peculiar spin-orbit hybridization, we find a noncollinear twisted RKKY interaction comprising XYZ-Heisenberg, symmetric in-plane, and asymmetric Dzyaloshinskii-Moriya (DM) terms. We see that contributions from the $z$ ($x$)-component of the XYZ-Heisenberg (DM) interaction are dominant for most parameters. The interactions, including DM terms that are responsible for interesting spin textures, require doping in most cases. We propose to modify the interactions in situ via optical control of band structure, and thereby doping. A notable aspect of this control protocol is breaking of electron-hole symmetry, which stems from the DM interaction.
△ Less
Submitted 20 December, 2022;
originally announced December 2022.
-
Multilingual Coreference Resolution in Multiparty Dialogue
Authors:
Boyuan Zheng,
Patrick Xia,
Mahsa Yarmohammadi,
Benjamin Van Durme
Abstract:
Existing multiparty dialogue datasets for entity coreference resolution are nascent, and many challenges are still unaddressed. We create a large-scale dataset, Multilingual Multiparty Coref (MMC), for this task based on TV transcripts. Due to the availability of gold-quality subtitles in multiple languages, we propose reusing the annotations to create silver coreference resolution data in other l…
▽ More
Existing multiparty dialogue datasets for entity coreference resolution are nascent, and many challenges are still unaddressed. We create a large-scale dataset, Multilingual Multiparty Coref (MMC), for this task based on TV transcripts. Due to the availability of gold-quality subtitles in multiple languages, we propose reusing the annotations to create silver coreference resolution data in other languages (Chinese and Farsi) via annotation projection. On the gold (English) data, off-the-shelf models perform relatively poorly on MMC, suggesting that MMC has broader coverage of multiparty coreference than prior datasets. On the silver data, we find success both using it for data augmentation and training from scratch, which effectively simulates the zero-shot cross-lingual setting.
△ Less
Submitted 8 July, 2023; v1 submitted 2 August, 2022;
originally announced August 2022.
-
Giant Resonant Self-Blocking in Magnetophononically Driven Quantum Magnets
Authors:
M. Yarmohammadi,
M. Krebs,
G. S. Uhrig,
B. Normand
Abstract:
Magnetophononics, the modulation of magnetic interactions by driven infrared-active lattice excitations, is emerging as a key mechanism for the ultrafast dynamical control of both semiclassical and quantum spin systems by coherent light. We demonstrate that, in a quantum magnetic material with strong spin-phonon coupling, resonances between the driven phonon and the spin excitation frequencies exh…
▽ More
Magnetophononics, the modulation of magnetic interactions by driven infrared-active lattice excitations, is emerging as a key mechanism for the ultrafast dynamical control of both semiclassical and quantum spin systems by coherent light. We demonstrate that, in a quantum magnetic material with strong spin-phonon coupling, resonances between the driven phonon and the spin excitation frequencies exhibit a giant self-blocking effect. Instead of absorbing more energy, the spin system acts as a strong brake on the driven phonon, causing it to absorb only a tiny fraction of the power available from the laser. Using the quantum master equations governing the nonequilibrium steady states of the coupled spin-lattice system, we show how self-blocking dominates the dynamics, demonstrate the creation of mutually repelling hybrid spin-phonon states, and control the nonequilibrium renormalization of the lattice-driven spin excitation band.
△ Less
Submitted 8 December, 2021;
originally announced December 2021.
-
Everything Is All It Takes: A Multipronged Strategy for Zero-Shot Cross-Lingual Information Extraction
Authors:
Mahsa Yarmohammadi,
Shijie Wu,
Marc Marone,
Haoran Xu,
Seth Ebner,
Guanghui Qin,
Yunmo Chen,
Jialiang Guo,
Craig Harman,
Kenton Murray,
Aaron Steven White,
Mark Dredze,
Benjamin Van Durme
Abstract:
Zero-shot cross-lingual information extraction (IE) describes the construction of an IE model for some target language, given existing annotations exclusively in some other language, typically English. While the advance of pretrained multilingual encoders suggests an easy optimism of "train on English, run on any language", we find through a thorough exploration and extension of techniques that a…
▽ More
Zero-shot cross-lingual information extraction (IE) describes the construction of an IE model for some target language, given existing annotations exclusively in some other language, typically English. While the advance of pretrained multilingual encoders suggests an easy optimism of "train on English, run on any language", we find through a thorough exploration and extension of techniques that a combination of approaches, both new and old, leads to better performance than any one cross-lingual strategy in particular. We explore techniques including data projection and self-training, and how different pretrained encoders impact them. We use English-to-Arabic IE as our initial example, demonstrating strong performance in this setting for event extraction, named entity recognition, part-of-speech tagging, and dependency parsing. We then apply data projection and self-training to three tasks across eight target languages. Because no single set of techniques performs the best across all tasks, we encourage practitioners to explore various configurations of the techniques described in this work when seeking to improve on zero-shot training.
△ Less
Submitted 14 September, 2021;
originally announced September 2021.
-
Dynamic mean-field theory for dense spin systems at infinite temperature
Authors:
Timo Gräßer,
Philip Bleicker,
Dag-Björn Hering,
Mohsen Yarmohammadi,
Götz S. Uhrig
Abstract:
A dynamic mean-field theory for spin ensembles (spinDMFT) at infinite temperatures on arbitrary lattices is established. The approach is introduced for an isotropic Heisenberg model with $S = \tfrac12$ and external field. For large coordination numbers, it is shown that the effect of the environment of each spin is captured by a classical time-dependent random mean-field which is normally distribu…
▽ More
A dynamic mean-field theory for spin ensembles (spinDMFT) at infinite temperatures on arbitrary lattices is established. The approach is introduced for an isotropic Heisenberg model with $S = \tfrac12$ and external field. For large coordination numbers, it is shown that the effect of the environment of each spin is captured by a classical time-dependent random mean-field which is normally distributed. Expectation values are calculated by averaging over these mean-fields, i.e., by a path integral over the normal distributions. A self-consistency condition is derived by linking the moments defining the normal distributions to spin autocorrelations. In this framework, we explicitly show how the rotating wave approximation becomes a valid description for increasing magnetic field. We also demonstrate that the approach can easily be extended. Exemplarily, we employ it to reach a quantitative understanding of a dense ensemble of spins with dipolar interaction which are distributed randomly on a plane including static Gaussian noise as well.
△ Less
Submitted 2 February, 2022; v1 submitted 16 July, 2021;
originally announced July 2021.
-
Gradual Fine-Tuning for Low-Resource Domain Adaptation
Authors:
Haoran Xu,
Seth Ebner,
Mahsa Yarmohammadi,
Aaron Steven White,
Benjamin Van Durme,
Kenton Murray
Abstract:
Fine-tuning is known to improve NLP models by adapting an initial model trained on more plentiful but less domain-salient examples to data in a target domain. Such domain adaptation is typically done using one stage of fine-tuning. We demonstrate that gradually fine-tuning in a multi-stage process can yield substantial further gains and can be applied without modifying the model or learning object…
▽ More
Fine-tuning is known to improve NLP models by adapting an initial model trained on more plentiful but less domain-salient examples to data in a target domain. Such domain adaptation is typically done using one stage of fine-tuning. We demonstrate that gradually fine-tuning in a multi-stage process can yield substantial further gains and can be applied without modifying the model or learning objective.
△ Less
Submitted 1 September, 2021; v1 submitted 3 March, 2021;
originally announced March 2021.
-
CopyNext: Explicit Span Copying and Alignment in Sequence to Sequence Models
Authors:
Abhinav Singh,
Patrick Xia,
Guanghui Qin,
Mahsa Yarmohammadi,
Benjamin Van Durme
Abstract:
Copy mechanisms are employed in sequence to sequence models (seq2seq) to generate reproductions of words from the input to the output. These frameworks, operating at the lexical type level, fail to provide an explicit alignment that records where each token was copied from. Further, they require contiguous token sequences from the input (spans) to be copied individually. We present a model with an…
▽ More
Copy mechanisms are employed in sequence to sequence models (seq2seq) to generate reproductions of words from the input to the output. These frameworks, operating at the lexical type level, fail to provide an explicit alignment that records where each token was copied from. Further, they require contiguous token sequences from the input (spans) to be copied individually. We present a model with an explicit token-level copy operation and extend it to copying entire spans. Our model provides hard alignments between spans in the input and output, allowing for nontraditional applications of seq2seq, like information extraction. We demonstrate the approach on Nested Named Entity Recognition, achieving near state-of-the-art accuracy with an order of magnitude increase in decoding speed.
△ Less
Submitted 28 October, 2020;
originally announced October 2020.
-
Dynamical properties of a driven dissipative dimerized $S = 1/2$ chain
Authors:
M. Yarmohammadi,
C. Meyer,
B. Fauseweh,
B. Normand,
G. S. Uhrig
Abstract:
We consider the dynamical properties of a gapped quantum spin system coupled to the electric field of a laser, which drives the resonant excitation of specific phonon modes that modulate the magnetic interactions. We deduce the quantum master equations governing the time-evolution of both the lattice and spin sectors, by developing a Lindblad formalism with bath operators providing an explicit des…
▽ More
We consider the dynamical properties of a gapped quantum spin system coupled to the electric field of a laser, which drives the resonant excitation of specific phonon modes that modulate the magnetic interactions. We deduce the quantum master equations governing the time-evolution of both the lattice and spin sectors, by developing a Lindblad formalism with bath operators providing an explicit description of their respective phonon-mediated damping terms. We investigate the nonequilibrium steady states (NESS) of the spin system established by a continuous driving, delineating parameter regimes in driving frequency, damping, and spin-phonon coupling for the establishment of physically meaningful NESS and their related non-trivial properties. Focusing on the regime of generic weak spin-phonon coupling, we characterize the NESS by their frequency and wave-vector content, explore their transient and relaxation behavior, and discuss the energy flow, the system temperature, and the critical role of the type of bath adopted. Our study lays a foundation for the quantitative modelling of experiments currently being designed to control coherent many-body spin states in quantum magnetic materials.
△ Less
Submitted 30 January, 2021; v1 submitted 30 September, 2020;
originally announced September 2020.
-
Bayesian Inference For Exponential Distribution Based On Upper Record Range
Authors:
P. Nasiri,
S. Hosseini,
M. Yarmohammadi,
F. Hatami
Abstract:
This paper deals with Bayesian estimations of scale parameter of the exponential distribution based on upper record range (Rn). This has been done in two steps; point and interval. In the first step the quadratic, squared error and absolute error, loss functions have been considered to obtain Bayesian-point estimations. Also in the next step the shortest Bayes interval (Hight Posterior Density int…
▽ More
This paper deals with Bayesian estimations of scale parameter of the exponential distribution based on upper record range (Rn). This has been done in two steps; point and interval. In the first step the quadratic, squared error and absolute error, loss functions have been considered to obtain Bayesian-point estimations. Also in the next step the shortest Bayes interval (Hight Posterior Density interval) and Bayes interval with equal tails based on upper record range have been found. Therefore, the Homotopy Perturbation Method(HPM) has been applied to obtain the limits of Hight Posterior Density intervals. Moreover, efforts have been made to meet the admissibility conditions for linear estimators based on upper record range of the form mRn+d by obtained Bayesian point estimations. So regarding the consideration of loss functions, the prior distribution between the conjunction family has been chosen to be able to produce the linear estimations from upper record range statistics. Finally, some numerical examples and simulations have been presented.
△ Less
Submitted 25 June, 2013;
originally announced June 2013.
