arXiv subjects

Jin Zhang

Publications and source records attributed to Jin Zhang.

At least 127 records · Page 7Linked to original sources

arXiv2025

Learning to Plan Before Answering: Self-Teaching LLMs to Learn Abstract Plans for Problem Solving

In the field of large language model (LLM) post-training, the effectiveness of utilizing synthetic data generated by the LLM itself has been well-presented. However, a key question remains unaddressed: what essential information should such self-generated data encapsulate? Existing approaches only produce step-by-step problem solutions, and fail to capture the abstract meta-knowledge necessary for generalization across similar problems. Drawing insights from cognitive science, where humans employ high-level abstraction to simplify complex problems before delving into specifics, we introduce a novel self-training algorithm: LEarning to Plan before Answering (LEPA). LEPA trains the LLM to formulate anticipatory plans, which serve as abstract meta-knowledge for problem-solving, before engaging with the intricacies of problems. This approach not only outlines the solution generation path but also shields the LLM from the distraction of irrelevant details. During data generation, LEPA first crafts an anticipatory plan based on the problem, and then generates a solution that aligns with both the plan and the problem. LEPA refines the plan through self-reflection, aiming to acquire plans that are instrumental in yielding correct solutions. During model optimization, the LLM is trained to predict both the refined plans and the corresponding solutions. By efficiently extracting and utilizing the anticipatory plans, LEPA demonstrates remarkable superiority over conventional algorithms on various challenging natural language reasoning benchmarks.

cs.CL↗

arXiv2025

Symmetry classification and invariant solutions of the classical geometric mean reversion process

Based on the Lie symmetry method, we investigate a Feynman-Kac formula for the classical geometric mean reversion process, which effectively describing the dynamics of short-term interest rates. The Lie algebra of infinitesimal symmetries and the corresponding one-parameter symmetry groups of the equation are obtained. An optimal system of invariant solutions are constructed by a derived optimal system of one-dimensional subalgebras. Because of taking into account a supply response to price rises, this equation provides for a more realistic assumption than the geometric Brownian motion in many investment scenarios.

math.DS↗

arXiv2025

Domain Generalization for Face Anti-spoofing via Content-aware Composite Prompt Engineering

The challenge of Domain Generalization (DG) in Face Anti-Spoofing (FAS) is the significant interference of domain-specific signals on subtle spoofing clues. Recently, some CLIP-based algorithms have been developed to alleviate this interference by adjusting the weights of visual classifiers. However, our analysis of this class-wise prompt engineering suffers from two shortcomings for DG FAS: (1) The categories of facial categories, such as real or spoof, have no semantics for the CLIP model, making it difficult to learn accurate category descriptions. (2) A single form of prompt cannot portray the various types of spoofing. In this work, instead of class-wise prompts, we propose a novel Content-aware Composite Prompt Engineering (CCPE) that generates instance-wise composite prompts, including both fixed template and learnable prompts. Specifically, our CCPE constructs content-aware prompts from two branches: (1) Inherent content prompt explicitly benefits from abundant transferred knowledge from the instruction-based Large Language Model (LLM). (2) Learnable content prompts implicitly extract the most informative visual content via Q-Former. Moreover, we design a Cross-Modal Guidance Module (CGM) that dynamically adjusts unimodal features for fusion to achieve better generalized FAS. Finally, our CCPE has been validated for its effectiveness in multiple cross-domain experiments and achieves state-of-the-art (SOTA) results.

cs.CV↗

arXiv2025

Auto-resolving the atomic structure at van der Waals interfaces using a generative model

The high-resolution visualization of atomic structures is significant for understanding the relationship between the microscopic configurations and macroscopic properties of materials. However, a rapid, accurate, and robust approach to automatically resolve complex patterns in atomic-resolution microscopy remains difficult to implement. Here, we present a Trident strategy-enhanced disentangled representation learning method (a generative model), which utilizes a few unlabelled experimental images with abundant low-cost simulated images to generate a large corpus of annotated simulation data that closely resembles experimental results, producing a high-quality large-volume training dataset. A structural inference model is then trained via a residual neural network which can directly deduce the interlayer slip and rotation of diversified and complicated stacking patterns at van der Waals interfaces with picometer-scale accuracy across various materials (e.g. MoS2, WS2, ReS2, ReSe2, and 1T'-MoTe2) with different layer numbers , demonstrating robustness to defects, imaging quality, and surface contaminations. The framework can also identify pattern transition interfaces, quantify subtle motif variations, and discriminate moire patterns that are difficult to distinguish in frequency domains. Finally, the high-throughput processing ability of our method provides insights into a van der Waals epitaxy mode where various thermodynamically favorable slip stackings can coexist.

cond-mat.mtrl-sci↗

arXiv2025

Photostriction Facilitates Relaxation of Lattice Distortion in Two-Dimensional Perovskites

The photostriction effect, a light-induced mechanical deformation in materials, originates from the intricate interplay between lattice structure and electronic excitation. In photovoltaic semiconductors, this effect plays a crucial role in shaping non-equilibrium structural responses, yet its fundamental mechanism remains elusive. Here, we uncover lattice expansion and structural reconfiguration in two-dimensional (2D) perovskites driven by photoinduced excitation using first-principles calculations. Our findings reveal that the photoinduced carriers lead to a substantial lattice expansion by about 2%. The expanded lattice facilitates strain relaxation with the amplitude of 20% by increasing interatomic distances and reducing internal stresses, thereby enhancing structural stability. The lattice dynamics can be systematically engineered through photodoping density, unveiling a new pathway to modulate light-matter interactions in 2D perovskites. These insights not only advance the understanding of optically driven structural dynamics but also offer a guiding principle for optimizing next-generation high-efficiency photovoltaic devices and optoelectronics.

cond-mat.mtrl-sci↗

arXiv2025

Phase-matching of high harmonic generation in twisted solids

High harmonic generation (HHG) in solids could enable attosecond and ultraviolet light sources with high compactness, great controllability and rich functions. However, the HHG process is accompanied by a quite large wavevector mismatch that is uncompensated by any traditional phase-matching method, directly limiting its energy conversion efficiency. Here, we propose an effective strategy for phase-matching of HHG with arbitrary harmonic orders in solids. Two flakes of solids with an interlayer twist induce a nonlinear optical phase that depends on the crystal symmetry, twist angle and harmonic order, which can be accurately designed to compensate for the phase mismatch in HHG. Guided by the twist-phase-matching theory, we achieved a record-high conversion efficiency of $~1.5\times10^{-5}$ for the fifth HHG in twisted hexagonal boron nitride crystals with a total thickness of only 1 $μm$. Our work establishes a foundation for developing ultrashort-wavelength and ultrafast-pulse laser sources in compact solid-state tabletop systems for fundamental and applied sciences.

physics.optics↗

arXiv2025

CoMT: A Novel Benchmark for Chain of Multi-modal Thought on Large Vision-Language Models

Large Vision-Language Models (LVLMs) have recently demonstrated amazing success in multi-modal tasks, including advancements in Multi-modal Chain-of-Thought (MCoT) reasoning. Despite these successes, current benchmarks still follow a traditional paradigm with multi-modal input and text-modal output, which leads to significant drawbacks such as missing visual operations and vague expressions. Motivated by this, we introduce a novel Chain of Multi-modal Thought (CoMT) benchmark to address these limitations. Different from the traditional MCoT benchmark, CoMT requires both multi-modal input and multi-modal reasoning output, aiming to mimic human-like reasoning that inherently integrates visual operation. Specifically, CoMT consists of four categories: (1) Visual Creation, (2) Visual Deletion, (3) Visual Update, and (4) Visual Selection to comprehensively explore complex visual operations and concise expression in real scenarios. We evaluate various LVLMs and strategies on CoMT, revealing some key insights into the capabilities and limitations of the current approaches. We hope that CoMT can inspire more research on introducing multi-modal generation into the reasoning process.

cs.CV↗

arXiv2025

Photoexcitation-induced Stacking Transition Assisted by Intralayer Reconstruction in Charge-Density-Wave Materials

Laser excitation has emerged as an effective tool for probing microscopic interactions and manipulating phases of matter. Among charge density wave (CDW) materials, 1T-TaS2 has garnered significant attention due to its diverse stacking orders and photoexcited responses. However, the mechanisms driving transitions among different stacking orders and the microscopic out-of-equilibrium dynamics remain unclear. We elucidate that photoexcitation can introduce interlayer stacking order transitions facilitated by laser-induced intralayer reconstruction in 1T-TaS2. Importantly, our finding reveals a novel pathway to introduce different phases through laser excitations, apparently distinct from thermally-induced phase transitions via interlayer sliding. In particular, photoexcitation is able to considerably change potential energy surfaces and evoke collective lattice dynamics. Consequently, the laser-induced intralayer reconstruction plays a crucial role in interlayer stacking-order transition, offering a new method to create exotic stackings and quantum phases. The exploration opens up great opportunities for manipulating CDW phases and electronic properties on the femtosecond timescale.

cond-mat.mtrl-sci↗

arXiv2025

Simulating Chemistry with Fermionic Optical Superlattices

We show that quantum number preserving Ansätze for variational optimization in quantum chemistry find an elegant mapping to ultracold fermions in optical superlattices. Using native Hubbard dynamics, trial ground states of molecular Hamiltonians can be prepared and their molecular energies measured in the lattice. The scheme requires local control over interactions and chemical potentials and global control over tunneling dynamics, but foregoes the need for optical tweezers, shuttling operations, or long-range interactions. We describe a complete compilation pipeline from the molecular Hamiltonian to the sequence of lattice operations, thus providing a concrete link between quantum simulation and chemistry. Our work enables the application of recent quantum algorithmic techniques, such as Double Factorization and quantum Tailored Coupled Cluster, to present-day fermionic optical lattice systems with significant improvements in the required number of experimental repetitions. We provide detailed quantum resource estimates for small non-trivial hardware experiments.

cond-mat.quant-gas↗

arXiv2025

qNBO: quasi-Newton Meets Bilevel Optimization

Bilevel optimization, addressing challenges in hierarchical learning tasks, has gained significant interest in machine learning. The practical implementation of the gradient descent method to bilevel optimization encounters computational hurdles, notably the computation of the exact lower-level solution and the inverse Hessian of the lower-level objective. Although these two aspects are inherently connected, existing methods typically handle them separately by solving the lower-level problem and a linear system for the inverse Hessian-vector product. In this paper, we introduce a general framework to address these computational challenges in a coordinated manner. Specifically, we leverage quasi-Newton algorithms to accelerate the resolution of the lower-level problem while efficiently approximating the inverse Hessian-vector product. Furthermore, by exploiting the superlinear convergence properties of BFGS, we establish the non-asymptotic convergence analysis of the BFGS adaptation within our framework. Numerical experiments demonstrate the comparable or superior performance of the proposed algorithms in real-world learning tasks, including hyperparameter optimization, data hyper-cleaning, and few-shot meta-learning.

cs.LG↗

arXiv2025

X-ray Emission Properties of a Compact Symmetric Object Sample

We present a comprehensive analysis of the X-ray observations obtained from \xmm\, and \chandra\, for a sample of bona-fide Compact Symmetric Objects (CSOs) to investigate their X-ray emission properties. Ultimately, we obtain 32 effective X-ray observational spectra from 17 CSOs. Most spectra can be well described by an absorbed single power-law model, with the exception of 6 spectra requiring an additional component in the soft X-ray band and 2 spectra exhibiting an iron emission line component. The data analysis results unveil the diverse characteristics of X-ray emission from CSOs. The sample covers X-ray luminosity ranging within $10^{40}-10^{45}$ erg s$^{-1}$, intrinsic absorbing column density ($N_{\rm H}^{\rm int}$) ranging within $10^{20}-10^{23}$ cm$^{-2}$, and photon spectral index ($Γ_{\rm X}$) ranging within 0.75--3.0. None of the CSOs in our sample have $N_{\rm H}^{\rm int}$ > $10^{23}\rm~cm^{-2}$, indicating that the X-ray emission in these CSOs is not highly obscured. The distribution of $Γ_{\rm X}$ for these CSOs closely resembles that observed in a sample of radio-loud quasars and low-excitation radio galaxies (RGs). In the radio--X-ray luminosity panel, these CSOs exhibit a distribution more akin to FR I RGs than FR II RGs, characterized by higher luminosities. The positive correlation between $Γ_{\rm X}$ and the Eddington ratio, which has been noted in radio-quiet active galactic nuclei, is not observed in these CSOs. These findings suggest that although the contribution of the disk-corona system cannot be completely ruled out, jet/lobe radiation likely plays a dominant role in the X-ray emission of these CSOs.

astro-ph.HE↗

arXiv2025

Nonequilibrium thermodynamic foundation of the grand-potential phase field model

Choosing the correct free energy functional is critical when developing thermodynamically consistent phase field models. We show that the grand-potential phase field model minimizes the Helmholtz free energy when mass conservation is imposed. While both functionals are at a minimum at equilibrium, the Helmholtz free energy decreases monotonically with time in the grand-potential phase field model, whereas the grand potential does not. Minimizing the grand potential implies a different problem where a system can exchange mass with its surroundings at every point, leading to a condition of isochemical potential and invalidating mass conservation of the system.

cond-mat.stat-mech↗

arXiv2025

Adaptive Sampled Softmax with Inverted Multi-Index: Methods, Theory and Applications

The softmax function is a cornerstone of multi-class classification, integral to a wide range of machine learning applications, from large-scale retrieval and ranking models to advanced large language models. However, its computational cost grows linearly with the number of classes, which becomes prohibitively expensive in scenarios with millions or even billions of classes. The sampled softmax, which relies on self-normalized importance sampling, has emerged as a powerful alternative, significantly reducing computational complexity. Yet, its estimator remains unbiased only when the sampling distribution matches the true softmax distribution. To improve both approximation accuracy and sampling efficiency, we propose the MIDX Sampler, a novel adaptive sampling strategy based on an inverted multi-index approach. Concretely, we decompose the softmax probability into several multinomial probabilities, each associated with a specific set of codewords and the last associated with the residual score of queries, thus reducing time complexity to the number of codewords instead of the number of classes. To further boost efficiency, we replace the query-specific residual probability with a simple uniform distribution, simplifying the computation while retaining high performance. Our method is backed by rigorous theoretical analysis, addressing key concerns such as sampling bias, gradient bias, convergence rates, and generalization error bounds. The results demonstrate that a smaller divergence from the ideal softmax distribution leads to faster convergence and improved generalization. Extensive experiments on large-scale language models, sequential recommenders, and extreme multi-class classification tasks confirm that the MIDX-Sampler delivers superior effectiveness and efficiency compared to existing approaches.

cs.LG↗

arXiv2025

A metamaterial telescope at millimetre wavelengths

In this paper we present a novel telescope composed exclusively of thin, flat optical elements, each being a hot-pressed multi-layered structure combining the properties of a lens, its anti-reflection coating and frequency selection or filtering. We discuss the design process, from fundamental physical metamaterial properties of the single periodic cell structure to the lens concept, which constitutes the building block of the telescope design, and the iterative process that is part of the lens optimization. We provide the results of a laboratory test campaign for different telescope designs based on three-lens arrangements. Beam cuts and focus measurements both on- and off-axis are compared with models showing good agreement. We conclude that a broad-band mm-wave complete telescope system consisting entirely of metamaterial flat lenses has been built and tested, showing comparable performance with conventional state-of-the-art refractive telescopes in the same wavelength region. This new broadband design, highly efficient at frequencies between 90 and 190 GHz, offers multiple advantages. These include a $> 80\%$ weight reduction, reduced issues tied to coating-survivability at cryogenic temperatures caused by differential contraction exacerbated by non-flat surfaces, as well as a reduction in the overall number of components and mechanical mounts.

astro-ph.IM↗

arXiv2024

Detection of an Orphan X-ray Flare from a Blazar Candidate EP240709a with Einstein Probe

Blazars are often observed to flare across multiple wavelengths. Orphan flares from blazars have been only detected a few times, providing an opportunity to understand the structure of the jet in the accreting system. We report a remarkable orphan X-ray flare from a blazar candidate EP240709a, detected by Einstein Probe (EP) in July 2024. The multi-band spectral properties and variability support EP240709a as a high-energy peaked BL Lacertae-type object. The flux in 0.5-10 keV increases by at least 28 times to the value of low state in 2020, with non-detection of remarkable flaring in other bands during the same period. EP240709a exhibits the harder-when-brighter tendency in the X-ray band during the orphan flare, while its infrared-optical spectra are featureless. We employ one-zone and two-zone leptonic synchrotron self-Compton models to perform the spectral energy distribution fitting. Detecting this rare orphan flare shows the potential of EP in discovering peculiar activities from AGN in high-cadence X-ray sky surveys.

astro-ph.HE↗

arXiv2024

Global track finding based on the Hough transform in the STCF detector

The proposed Super Tau-Charm Facility (STCF) is an electron-positron collider designed to operate in a center-of-mass energy range from 2 to 7 GeV. It provides a unique platform for physics research in the tau-charm energy region. To fulfill the physics goals of STCF, high tracking efficiency and good momentum resolution is required for charged particles with momenta from 50 MeV/c to 3.5 GeV/c. A global track finding algorithm based on Hough transform has been developed and implemented in the STCF software framework to meet this requirement. The design of the algorithm and its performance with simulation are presented in this paper.

hep-ex↗

arXiv2024

Phase diagram of Rydberg atoms in a two-leg rectangular ladder

Using the density matrix renormalization group algorithm, we map the ground-state phase diagram of a two-leg Rydberg ladder array with lattice spacings $a_x=2a_y$. We identify various density wave phases that spontaneously break the translational symmetry or the top-bottom reflection symmetry within the ladder. By increasing the laser detuning from zero, where the system is in a disordered phase that preserves all symmetries, we observe density wave orders with spontaneous breaking of the translational $\mathbb{Z}_p$ symmetries at intermediate detuning values, while the reflection symmetry is preserved. These orders exhibit nonzero bond orders with positive expectation values on every $p$th rung, thus labeled as $\mathbb{Z}_p^+$ phases. At larger detuning values, another spontaneous breaking of the reflection symmetry, which disrupted the bond orders on the rungs, occurs via an Ising phase transition. In these phases, either the top or the bottom site is occupied in a staggered way on every $p$th rung, breaking the translational $\mathbb{Z}_{2p}$ symmetry, thus labeled by $\mathbb{Z}_{2p}$ phases. We locate and characterize the 3-state Potts point and Ashkin-Teller point along the commensurate lines, as well as the direct chiral phase transitions between the disordered phase and the $\mathbb{Z}_p^+$ ($p = 3, 4$) phases. Critical exponents $ν$ and $z$ are calculated for both conformal and chiral phase transition points. We finally identify two types of floating phases in the phase diagram: one characterized by a quasi-long-range incommensurate bond-order wave, and the other by a quasi-long-range incommensurate wave of density differences in the rungs. Our work motivates further applications of Rydberg atom arrays in quantum simulation.

cond-mat.quant-gas↗

arXiv2024

Ultrafast Polarization Switching via Laser-activated Ionic Migration in Ferroelectric CuInP$_2$S$_6$

As a layered ferroelectric material, CuInP2S6 has garnered significant attention for its robust ferroelectric state and potential applications in memory devices. In this work, we demonstrate that with short laser pulses ultrafast reversible polarization switching within hundreds of femtoseconds can be achieved in ferroelectric CuInP$_2$S$_6$. Specifically, photoexcitation triggers collective ionic migration and ferroelectricity reversal in CuInP$_2$S$_6$, revealing a novel pathway to access different ferroelectric phases through optical excitation. Our findings indicate that laser pulses substantially alter the transition barriers, promoting ionic transport facilitated by the photodoping effect. This laser-induced ionic migration proves critical for enabling polarization transitions, offering a novel pathway to explore and control exotic quantum phases. These insights open exciting possibilities for manipulating ferroelectric states and electronic properties on an unprecedented ultrafast timescale.

cond-mat.mtrl-sci↗