arXiv subjects

Mingda Li

Publications and source records attributed to Mingda Li.

At least 19 recordsLinked to original sources

arXiv2026

Agentic inverse design of spatially programmed materials

Spatially programmed functional materials, from functionally graded alloys and metamaterials to quantum superlattices, use their spatial structure to tailor physical properties. However, translating a desired spatial structure into a practical fabrication recipe remains challenging due to the complex relationship between processing conditions and resulting material structures. Here we show that a large language model (LLM) agent can inverse-design such fabrication recipes by using a physics-based simulator of ion-matter interactions as a tool. Given an arbitrary depth profile of implanted ions or vacancies, the agent selects the ion species, energies, incident angles, and target material layers; writes, runs, and interprets simulations; and iteratively refines the recipe until the simulated profile matches the target, with minimal human input. Our work demonstrates physics-guided agentic inverse design as a practical approach to making the spatial structure of materials a programmable variable, opening new opportunities to explore emergent physical phenomena and functional phases of matter.

cond-mat.mtrl-sci↗

arXiv2026

Autonomous Quantum Measurement with Agentic AI

Quantum materials exhibit diverse electronic and functional phenomena, but their characterization requires extensive experimental expertise, iterative decision-making and accumulated laboratory knowledge that is difficult to document and transfer. Here we introduce Quantum Agent (QUANTA), an agentic artificial intelligence (AI) framework for autonomous quantum-materials characterization. Using a widely adopted Physical Property Measurement System (PPMS) as an experimental platform, QUANTA combines retrieval augmentation using scientific literature, experimental protocols and prior measurements with specialized agents for experimental planning, troubleshooting, and data interpretation. A separate deterministic control layer validates instrument actions. Guided by incoming data, QUANTA evaluates measurement quality, revises experimental plans and selects subsequent measurements. We demonstrate these capabilities through transport measurements of the topological Weyl semimetal TaP, as well as investigations of nonlinear responses in magnetic material Co2MnGa. These results establish QUANTA as a generalizable framework for evidence-driven, safeguarded and closed-loop quantum-materials characterization.

cond-mat.mtrl-sci↗

arXiv2026

JANUS: A Multi-modal Foundation Neural Sampler for Disordered Materials

Disordered materials are central to modern technologies, from high-performance alloys to quantum devices, yet understanding and designing them requires exploring an enormous space of compositions and atomic arrangements. Here we introduce JANUS, a multimodal neural sampler that jointly explores chemical composition, atomic structure and volume, learning directly from the energy of the system without requiring pre-generated equilibrium configurations. Across test systems, JANUS reproduces reference equilibrium properties and recovers free energy and phase behavior with about three orders of magnitude fewer energy evaluations than reference Monte Carlo sampling. In multicomponent alloys, JANUS reproduces experimentally observed local chemical order and can be steered to further enhance chemical order and stiffness. Coupled to an evolutionary language model agent, it also enables efficient design of alloys with balanced optical and mechanical properties. Beyond alloys, JANUS explores vacancies and dopants spanning 15 elements in semiconductors including silicon and diamond, recovers established defects including the silicon $E$ centre, and identifies new candidate defect pairs and triplets for quantum engineering, including S-Ti in silicon and B-[O]-O in diamond, with deep in-gap states validated by hybrid-functional density functional theory. By unifying exploration across composition, atomic arrangement and volume, JANUS opens a new avenue for understanding, characterizing and designing chemically disordered materials.

cond-mat.mtrl-sci↗

arXiv2026

APM-Bench: Benchmarking Cross-session Persistent Memory for Egocentric Streaming Video Assistants

To serve as real-world personal assistants, streaming video models need persistent memory that retains past experiences for later use. Yet existing streaming benchmarks and methods often focus on individual continuous videos or short clips, overlooking that real-world interactions are often intermittent and require memory to persist across interruptions. To fill this gap, we introduce APM-Bench, which reformulates real-world streaming interaction as multi-session life trajectories. It contains 549 sessions, 104 trajectories, and 2,719 candidates, spanning both objective and open-ended questions. Each session is a video with fine-grained annotations, and sessions within a trajectory revolve around related activities. Models then use persistent memory to answer questions about past sessions and provide proactive responses while maintaining real-time interaction. This raises challenges: persistent memory must be storable, selectively retain information, be injected at the right time, and remain efficient. Moreover, finite storage may leave required evidence unavailable, so assistants should recognize missing evidence. Therefore, we systematically evaluate general video models under different memory protocols and diverse specialized streaming memory systems, and test whether models acknowledge insufficient evidence. Our evaluation reveals a clear utility--latency--storage trade-off: existing methods still struggle to simultaneously achieve reliable long-term recall, low overhead, and effective proactive assistance across sessions. APM-Bench provides a comprehensive testbed for developing and comparing persistent memory systems under realistic streaming conditions. We hope it encourages future work that jointly considers utility, latency, and storage toward more practical persistent memory for real-world streaming assistants.

cs.CV↗

arXiv2026

ATLAS: A Foundation Neural Sampler for Amorphous Materials

Amorphous materials exhibit exceptional mechanical and functional properties, yet their rugged energy landscapes are notoriously difficult to sample. Below the glass-transition temperature, conventional molecular dynamics and Monte Carlo become inefficient because equilibration relies on rare barrier-crossing events, while data-driven generative models are constrained by scarce and biased reference ensembles. Here, we introduce ATLAS, an efficient sampler that learns a diffusion process to generate Boltzmann-distributed amorphous structures directly from a target energy function. Parameterized by an equivariant graph neural network, ATLAS generalizes across system size, temperature, and composition. By exploiting the time reversal of the diffusion process, it enables efficient estimation of thermodynamic quantities and steering toward target observables. In two-dimensional Kob-Andersen systems, ATLAS reproduces parallel tempering Markov chain Monte Carlo structural distributions, free energies and entropies, achieving below 0.2% free energy error in the low-temperature glass regime with over 500-fold fewer energy evaluations. In Cu-Zr and Cr-Co-Ni metallic glasses, ATLAS recovers experimentally observed short-range-order trends and steers structures toward prescribed order parameters and optimized bulk moduli. Moreover, composition-amortized pretraining outperforms composition-specific training from scratch, reduces inverse-design costs by several hundred-fold, and enables sampling with expensive universal machine learning interatomic potentials. Coupled to a large language model agent, ATLAS searches an eight-element space for high-entropy metallic glasses balancing stiffness and ductility, identifying a converged Pareto frontier within 480 oracle evaluations. Together, these results establish ATLAS as a foundation model for sampling, steering and designing amorphous materials.

cond-mat.mtrl-sci↗

arXiv2026

A Polynomial-Scaling PDE Solver with Entanglement-Basis Tensor Networks

We develop a finite element method (FEM) for partial differential equation (PDE) solver based on the entanglement-basis representation introduced in our companion work. By lifting non-linear finite-element equations into an augmented coefficient space, the governing PDE together with boundary, initial, and inter-element constraints can be expressed through a unified quadratic residual minimization. Although this augmented space grows exponentially with the number of elements, its tensor-product structure allows it to be represented efficiently using tensor networks. Using the matrix product state (MPS) as a concrete example, we show that density matrix renormalization group (DMRG) sweeps enable element-by-element optimization without explicitly constructing the full augmented space. For bounded bond dimension, the resulting computational cost scales polynomially with the number of finite elements. We extend the framework to time-dependent problems through implicit temporal discretization and demonstrate convergence under both mesh and polynomial refinement using diffusion equations.

math.NA↗

arXiv2026

MARCO: Click-Intent Decomposition for Calibrated Ads Conversion Prediction

Not all clicks are equal. Industrial ads ranking decouples conversion probability into click-through rate (CTR) and post-click conversion rate (CVR), yet treats every click as the same event. In reality, users provide a free, self-generated signal of intent through their physical UI interactions. Different click types on the same ad exhibit a 4-fold difference in actual conversion rates. By conflating these signals, the standard CVR model under-predicts high-intent clicks and over-predicts low-intent ones, which is a bias masked by near-perfect aggregate calibration. We propose MARCO (Multi-intent Ads Ranking Composition Optimization), a framework that resolves this bias by decomposing each click by intent. Using the logged click type as a free behavioral label, MARCO trains per-intent CVR heads on homogeneous populations, and at serving time composes their per-intent CVR estimates under a predicted distribution over intents. Theoretically, we prove that decomposition never raises population risk, give the exact headroom under squared loss and non-negativity under the deployed loss, and show through a routing-efficiency dial how much of it reaches serving. Because the population-optimal score is unchanged, any gain is a finite-capacity estimation and calibration effect that we validated both offline and online. For deployment at scale, we further cast multi-impression, multi-click attribution as credit assignment with a bias-variance tradeoff analogous to RL return estimation, showing last-impression, first-click attribution is the low-bias, low-variance, deterministic choice under production constraints, and derive three consistency conditions enforced end-to-end at scale. Deployed at binary intent granularity, MARCO corrects per-intent calibration to approximately 100%, lifts conversions per click by +2.80%, and drives +0.98% cumulative improvement in topline metrics.

cs.LG↗

arXiv2026

LUT: Latent Utility Training for Visual Reasoning

Multimodal large language models have advanced visual understanding, yet perception-intensive reasoning remains challenging. Recent latent visual reasoning methods introduce hidden-space computation before answering, but they often rely on costly intermediate supervision, such as bounding boxes, sketches, or interleaved rationales. These strategies focus on how latent states should be shaped, but do not explicitly assess whether the latent is useful for the final answer. We propose LUT, a latent reasoning framework trained with only standard VQA pairs. LUT centers training on Latent Utility at two levels. At the trajectory level, we propose Utility-Aware Latent Distillation SFT, which explores answer-relevant latent trajectories, selects qualified trajectories by their information gain, and distills more reliable and learnable supervision through curriculum learning. At the step level, we propose Latent Attribution Policy Optimization, which uses answer-to-latent attribution to differentially optimize latent steps during reinforcement learning. Experiments on perception-intensive visual reasoning benchmarks show that LUT outperforms previous latent reasoning methods and remains competitive with latent-text interleaved methods with lower annotation cost.

cs.CV↗

arXiv2026

Agentic AI for Scientific Reasoning in Autonomous Quantum Sensing Experiments

We implement an agentic AI workflow built around a large language model (LLM) agent for autonomous experiments with nitrogen-vacancy (NV) centers in diamond. NV centers are a widely used platform for quantum sensing, and the ability to control many measurements from a computer makes NV experiments a natural setting for autonomous workflows. We make two main contributions. First, we demonstrate an autonomous NV experiment workflow that combines persistent project records, quantitative calculation and data analysis tools, and deterministic experiment control. In one autonomous experiment, the agent selected a single NV center, calibrated its resonant frequency, measured $T_2^\ast$ with Ramsey measurements, and added a Carr--Purcell--Meiboom--Gill (CPMG) measurement to check a weak feature that could be related to nearby $^{13}\mathrm{C}$. Second, we introduce two offline benchmarks that evaluate the agent's reasoning separately from laboratory execution. We evaluated both benchmarks with GPT-5.4, GPT-5.5, and GPT-5.6 Sol. In the Ramsey checkpoint benchmark, greater reasoning effort generally improved recognition of a residual resonance calibration offset. By contrast, in the pulsed optically detected magnetic resonance (pODMR) data evaluation benchmark, pulse sequence information alone produced more false positive resonance judgments at higher reasoning effort. Requiring an expected signal calculation kept false positive rates low across all three models and reasoning settings. The results suggest a clear division of labor for autonomous experiments. The agent forms scientific hypotheses and uses quantitative tools to evaluate data, while deterministic code controls the hardware and enforces safety constraints.

quant-ph↗

arXiv2026

Order of Magnitude Improved Optical Trapping of Molecules Through Transverse Cooling

We demonstrate a two-dimensional Sisyphus laser cooling method that increases the number of strontium monohydroxide (SrOH) molecules loaded into a magneto-optical trap by a factor of 12. Subsequent loading into an optical dipole trap (ODT) achieves $2.2 (3)\times10^4$ ultracold SrOH molecules with a peak density of $\sim2(1)\times10^{10}~\mathrm{cm^{-3}}$. The lifetime of molecules in the ODT is limited by two-body collisions characterized by a measured collision rate constant $β\sim 4\times10^{-10}~\mathrm{cm^3/s}$. The cooling method developed here is generally applicable to all known cases of direct molecular laser cooling, including symmetric and asymmetric top molecules. Increases in trapped molecule number will directly improve the search for ultralight dark matter, position polyatomic molecules as a platform for probing CP-violating new particles with masses $\gg$10 TeV, and facilitate a broad range of further research in quantum science.

physics.atom-ph↗

arXiv2026

Tensor-Network Finite Elements for Analytic Operator Equations

Operator equations (OEs) underpin quantitative modeling across science and engineering. Finite-element (FE) methods discretize continuous OEs into finite-dimensional algebraic systems, whereas tensor networks (TNs) provide flexible variational representations of correlated discrete systems. Here, we develop a framework that connects FE with TN for analytic OEs. The power of this method comes from its ability to convert highly non-linear partial differential equations into linear matrix equations. In particular, we show that FE discretization induces a hierarchy of multilinear interaction tensors, through which differential, integral, nonlinear, memory, and delay equations can be expressed within a common algebraic structure. The resulting systems are reformulated as weighted-residual optimization problems over TN degrees of freedom. Matrix-product-state calculations for one-dimensional linear and nonlinear diffusion reproduce conventional solutions with controlled error while preserving continuity and Neumann boundary conditions. The framework provides a common variational language for analytic OEs and establishes a direct connection between FE numerical formalism and TN variational algorithms, offering a general foundation for TN-based and quantum-inspired approaches to solving OEs.

math.NA↗

arXiv2026

XRDiff: Crystal Structure Prediction from Powder X-Ray Diffraction Data Using Diffusion Models

Determining the crystal structure of a material from its powder X-ray diffraction (PXRD) pattern is a central challenge in materials science. PXRD is an accessible and widely used characterization technique, yet recovering the atomic structure from diffraction data requires solving an underdetermined inverse problem due to the loss of phase information. Generative modeling can provide a prior over atomic structure and learn the mapping from PXRD patterns to crystal structures via simulated structure-spectrum pairs. We present XRDiff, a diffusion model that recovers crystal structures from PXRD given either the stoichiometry or, in a more challenging setting, the elemental constituents and total number of atoms in the unit cell. We evaluate on datasets where each stoichiometry has multiple polymorphs and all polymorphs of a given composition are held out together, ensuring that high performance reflects genuine use of the diffraction signal. XRDiff achieves strong structure recovery rates on simulated benchmarks, indicating that the model learns a spectrum-to-structure mapping precise enough to differentiate between polymorphs. To address generalization to experimental data, we compare a full-spectrum encoding against an encoding based on peak descriptors. The peak-based encoding generalizes substantially better, outperforming even a model trained on full spectra with augmentations fitted to the experimental noise distribution. These results demonstrate that representations robust to the noise and artifacts present in real-world PXRD offer a practical and scalable path toward closing the simulation-to-experiment gap, enabling zero-shot crystal structure solution from experimental PXRD with full or partial chemical composition input.

cond-mat.mtrl-sci↗

arXiv2026

Gradients with Respect to Semantics Preserving Embeddings Tell the Uncertainty of Large Language Models

Uncertainty quantification (UQ) is an important technique for ensuring the trustworthiness of LLMs, given their tendency to hallucinate. Existing state-of-the-art UQ approaches for free-form generation rely heavily on sampling, which incurs high computational cost and variance. In this work, we propose the first gradient-based UQ method for free-form generation, SemGrad, which is sampling-free and computationally efficient. Unlike prior gradient-based methods developed for classification tasks that operates in parameter space, we propose to consider gradients in semantic space. Our method builds on the key intuition that a confident LLM should maintain stable output distributions under semantically equivalent input perturbations. We interpret the stability as the gradients in semantic space and introduce a Semantic Preservation Score (SPS) to identify embeddings that best capture semantics, with respect to which gradients are computed. We further propose HybridGrad, which combines the strengths of SemGrad and parameter gradients. Experiments demonstrate that both of our methods provide efficient and effective uncertainty estimates, achieving superior performance than state-of-the-art methods, particularly in settings with multiple valid responses.

cs.CL↗

arXiv2026

Can LLMs extract scientific consensus? A case study in high-temperature superconductivity

Scientific knowledge is increasingly dispersed across vast and heterogeneous scientific literature, where important claims are often implicit, evolving, and internally debated. While large language models (LLMs) have shown impressive performance in information extraction and summarization, their ability to recover latent scientific consensus remains unclear. Here, we investigate this problem in the context of high-temperature superconductivity (HTS), a long-standing and highly debated topic in condensed matter physics, as a challenging testbed. Using near 18,000 highly-cited publications over the past seven decades, we construct a structured knowledge graph linking competing superconducting mechanisms, material families, evidential modalities, and citation relations. We find that LLM-extracted representations recover coherent and physically interpretable structures, including family-dependent mechanism profiles, evidence-specific correlations, and citation-mediated temporal evolution of scientific beliefs. Ablation studies on LLM further show that the global structure remains robust across prompting, decoding, and model variations. Our results suggest that LLMs can indeed serve as scalable tools for deciphering scientific knowledge in domains characterized by competing interpretations and evolving knowledge.

cs.DL↗

arXiv2026

Large Transverse Thermoelectric Effect in Weyl Semimetal TaIrTe$_4$ Engineered for Photodetection

Anomalous local photocurrent generation via second-order nonlinear and thermoelectric responses is a signature of many topological semimetals. The emergence of these photocurrents is inherently linked to symmetry breaking and anisotropy of their crystal lattices. Studies of type-II Weyl semimetals of group C$_{2v}$ (WTe$_2$, MoTe$_2$, TaIrTe$_4$) have reported anomalous, nonlocal photocurrents localized to crystals edges or far from electrodes, which are highly dependent on the geometry of the material sample. While originally attributed to a nonlinear charge current response, it was recently shown that these currents could instead be attributed to the anisotropic Seebeck coefficients of the materials. Here, we confirm that anomalous photocurrents observed in TaIrTe$_4$ under either visible or far-infrared far-field illumination originate from the large transverse thermoelectric effect. We engineer the mutual orientation of crystal edges and electrodes as well as the thermal environment of TaIrTe$_4$ to control and amplify its spatial photocurrent response. We show that substrate engineering can locally enhance photocurrent. This framework of thermal device engineering can enable broadband photo detection schemes by leveraging spectral and spatial dependence of photocurrents for applications like wavefront sensing, beam positioning, and edge detection.

cond-mat.mtrl-sci↗

arXiv2026

Universal Magnetic Structure Prediction from Atomic Coordinates with Near-Experimental Accuracy

Magnetic order is a fundamental property of materials, governing collective behavior and enabling a broad range of functionalities. Yet magnetic structure remains difficult to determine: experiments are costly and specialized, while first-principles methods often struggle with the noncollinear and incommensurate orders found in real materials. Here we introduce magnetic structure network (MSN), an E(3) equivariant graph neural network that predicts both collinear and non-collinear magnetic structures directly from atomic crystal structures, trained directly on experimentally determined structures from MAGNDATA. By proposing the primitive modulated structure representation (PMSR), we are able to encode commensurate and incommensurate structures in a unified way without symmetry assumptions. The model achieves strong performance across all modulation components and reconstructs experimental magnetic structures with high fidelity. Our approach provides a scalable framework for rapid magnetic structure prediction and opens a route to data-driven discovery of magnetic materials.

cond-mat.mtrl-sci↗

arXiv2026

Probing Non-Equilibrium Grain Boundary Dynamics with XPCS and Domain-Adaptive Machine Learning

Grain-boundary (GB) dynamics control the stability, mechanical, and functional response of nanocrystalline materials, but direct experimental access to their slow non-equilibrium motion has been limited. Here we establish X-ray photon correlation spectroscopy (XPCS), combined with domain-adaptive machine learning, as a quantitative probe of GB dynamics. Temperature- and grain-size-dependent two-time XPCS measurements in nanocrystalline silicon reveal pronounced departures from time-translation invariance, showing that GB relaxation can remain far from equilibrium over experimental timescales. However, direct extraction of quantitative physical information from these high-dimensional, noisy fluctuation maps faces a significant challenge. To overcome this barrier, we develop a semi-supervised learning framework that transfers physical parameter labels from continuum simulations to unlabeled experimental XPCS maps through domain-adaptive representation alignment. This AI-augmented approach enables the extraction of key kinetic parameters, including bulk diffusivity, GB stiffness, and effective GB concentration, directly from experimental XPCS measurements. Our results show how machine learning can transform indirect fluctuation signals into quantitative materials dynamics, providing a general route to study non-equilibrium defect motion in solids.

cond-mat.mtrl-sci↗

arXiv2026

Quantum Theory of Functionally Graded Materials

Functionally graded materials (FGMs) are composites whose composition or microstructure varies continuously in space, producing position-dependent mechanical and functional properties. In recent years, FGMs have gained significant attention due to advances in additive manufacturing, which enable precise spatial control of composition and orientation. However, their graded, aperiodic structure breaks the assumptions of Bloch's theorem, making first-principles electronic and electromagnetic calculations challenging. Here we develop an ab initio quantum theoretical framework for the electromagnetic properties of FGMs. Using a non-interacting electron model, we formulate a theory of modulated Bloch states, derive effective field equations, and solve them by proposing a generalized WKB (GWKB) method, an effective mass approximation, the Boltzmann equation, and numerical approaches. Our GWKB solution is not semiclassical but remains valid in the fully quantum regime. We show that effective observables such as conductivity, magnetic permeability, and electric permittivity generally do not admit a tensorial description in graded media, and that engineered orientational gradients enable precise control of Landau quantization. As a device example, we further develop a theory of graded p-n junctions with enhanced electronic tunability. This framework lays the quantum foundation for predictive design of graded composite materials, enabling AI-accelerated discovery of next-generation functional architectures.

cond-mat.mtrl-sci↗