arXiv subjects

Yi Jiang

Publications and source records attributed to Yi Jiang.

At least 91 records · Page 5Linked to original sources

arXiv2025

Mic-hackathon 2024: Hackathon on Machine Learning for Electron and Scanning Probe Microscopy

Microscopy is a primary source of information on materials structure and functionality at nanometer and atomic scales. The data generated is often well-structured, enriched with metadata and sample histories, though not always consistent in detail or format. The adoption of Data Management Plans (DMPs) by major funding agencies promotes preservation and access. However, deriving insights remains difficult due to the lack of standardized code ecosystems, benchmarks, and integration strategies. As a result, data usage is inefficient and analysis time is extensive. In addition to post-acquisition analysis, new APIs from major microscope manufacturers enable real-time, ML-based analytics for automated decision-making and ML-agent-controlled microscope operation. Yet, a gap remains between the ML and microscopy communities, limiting the impact of these methods on physics, materials discovery, and optimization. Hackathons help bridge this divide by fostering collaboration between ML researchers and microscopy experts. They encourage the development of novel solutions that apply ML to microscopy, while preparing a future workforce for instrumentation, materials science, and applied ML. This hackathon produced benchmark datasets and digital twins of microscopes to support community growth and standardized workflows. All related code is available at GitHub: https://github.com/KalininGroup/Mic-hackathon-2024-codes-publication/tree/1.0.0.1

cond-mat.mtrl-sci↗

arXiv2025

Challenging Dataset and Multi-modal Gated Mixture of Experts Model for Remote Sensing Copy-Move Forgery Understanding

The Remote Sensing Copy-Move Question Answering (RSCMQA) task focuses on interpreting complex tampering scenarios and inferring the relationships between objects. Currently, publicly available datasets often use randomly generated tampered images, which lack spatial logic and do not meet the practical needs of defense security and land resource monitoring. To address this, we propose a high-quality manually annotated RSCMQA dataset, Real-RSCM, which provides more realistic evaluation metrics for the identification and understanding of remote sensing image tampering. The tampered images in the Real-RSCM dataset are subtle, authentic, and challenging, posing significant difficulties for model discrimination capabilities. To overcome these challenges, we introduce a multimodal gated mixture of experts model (CM-MMoE), which guides multi-expert models to discern tampered information in images through multi-level visual semantics and textual joint modeling. Extensive experiments demonstrate that CM-MMoE provides a stronger benchmark for the RSCMQA task compared to general VQA and CMQA models. Our dataset and code are available at https://github.com/shenyedepisa/CM-MMoE.

cs.MM↗

arXiv2025

Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis

We present Infinity, a Bitwise Visual AutoRegressive Modeling capable of generating high-resolution, photorealistic images following language instruction. Infinity redefines visual autoregressive model under a bitwise token prediction framework with an infinite-vocabulary tokenizer & classifier and bitwise self-correction mechanism, remarkably improving the generation capacity and details. By theoretically scaling the tokenizer vocabulary size to infinity and concurrently scaling the transformer size, our method significantly unleashes powerful scaling capabilities compared to vanilla VAR. Infinity sets a new record for autoregressive text-to-image models, outperforming top-tier diffusion models like SD3-Medium and SDXL. Notably, Infinity surpasses SD3-Medium by improving the GenEval benchmark score from 0.62 to 0.73 and the ImageReward benchmark score from 0.87 to 0.96, achieving a win rate of 66%. Without extra optimization, Infinity generates a high-quality 1024x1024 image in 0.8 seconds, making it 2.6x faster than SD3-Medium and establishing it as the fastest text-to-image model. Models and codes will be released to promote further exploration of Infinity for visual generation and unified tokenizer modeling.

cs.CV↗

arXiv2025

CeCo$_2$P$_2$: a unique Co-antiferromagnetic topological heavy-fermion system with $P\cdot T$-protected Kondo effect and nodal-line excitations

Based on high-throughput screening and experimental data, we find that CeCo$_2$P$_2$ is unique in heavy-fermion materials: it has a Kondo effect at a high temperature which is nonetheless below a Co-antiferromagnetic ordering temperature. This begs the question: how is the Kondo singlet formed? \emph{All} other magnetic Kondo materials do not first form magnetism on the atoms whose electrons are supposed to screen the local moments. We theoretically explain these observations and show the multifaceted uniqueness of CeCo$_2$P$_2$: a playground for Kondo, magnetism, flat band, and topological physics. At high temperatures, the itinerant Co $c$ electrons of the system form non-atomic bands with a narrow bandwidth, leading to a high antiferromagnetic transition temperature. We show that the quantum geometry of the bands promotes in-plane ferromagnetism, while the weak dispersion along the $z$ direction facilitates out-of-plane antiferromagnetism. At low temperatures, we uncover a novel phase that manifests the coexistence of Co-antiferromagnetism and the Kondo effect, linked to the $P\cdot T$-protected Kramers' doublets and the filling-enforced metallic nature of $c$ electrons in the antiferromagnetic phase. Subsequently, the emergence of the Kondo effect, in cooperation with glide-mirror-$z$ symmetry, creates nodal-line excitation near the Fermi energy. Our results emphasize the importance of lattice symmetry and quantum geometry, Kondo physics, and magnetism in the understanding of the correlation physics of this unique compound. We also test our theory on the structurally similar compound LaCo$_2$P$_2$ and show how we are able to understand its vastly different phase diagram.

cond-mat.str-el↗

arXiv2025

Topological Invariants in Nonlinear Thouless Pumping of Solitons

Recent explorations of quantized solitons transport in optical waveguides have thrust nonlinear topological pumping into the spotlight. In this work, we introduce a unified topological invariant applicable across both weakly and strongly nonlinear regimes. In the weak nonlinearity regime, where the nonlinear bands are wellseparated, the invariant reduces to the Abelian Chern number of the occupied nonlinear band. Consequently, the pumped charge is quantized to an integer value. As the nonlinearity increases, the nonlinear bands start to intertwine, leading to a situation where the invariant is expressed as the non-Abelian Chern number divided by the number of interacting bands. This could result in a fractional quantization of the pumped charge. Our unified topological invariant approach not only advances the understanding of the soliton dynamics, but also provides implications for the future design of nonlinear topological systems.

physics.atom-ph↗

arXiv2025

Direct Joule-Heated Non-Equilibrium Synthesis Enables High Performing Thermoelectrics

High-throughput synthesis of bulk inorganic materials is crucial for accelerating functional materials discovery but is hindered by slow, energy-intensive solid-state methods. We introduce Direct Joule-Heated Synthesis (DJS), a rapid, single-step and scalable solid-state synthesis technique achieving a $10^5$-fold speedup and 20,000x energy efficiency improvement over conventional synthesis. DJS enables the synthesis of dense, bulk chalcogenides ($\mathrm{Bi_{0.5}Sb_{1.5}Te_3}$, $\mathrm{AgSbTe_2}$), achieving a zT of 2.3 at 573 K in optimally Cd/Se co-doped $\mathrm{AgSbTe_2}$, one of the highest for polycrystalline materials at this temperature. DJS enables optimal co-doping and rapid, non-equilibrium solidification, producing lamellar microstructures, interfacial regions, and cation-ordered nanodomains that scatter all-scale phonons, achieving ultralow lattice thermal conductivity (~0.2 $W m^{-1} K^{-1}$ at 573 K). DJS establishes a new benchmark for scalable and fast synthesis, accelerating functional material discovery.

cond-mat.mtrl-sci↗

arXiv2025

GainRAG: Preference Alignment in Retrieval-Augmented Generation through Gain Signal Synthesis

The Retrieval-Augmented Generation (RAG) framework introduces a retrieval module to dynamically inject retrieved information into the input context of large language models (LLMs), and has demonstrated significant success in various NLP tasks. However, the current study points out that there is a preference gap between retrievers and LLMs in the RAG framework, which limit the further improvement of system performance. Some highly relevant passages may interfere with LLM reasoning because they contain complex or contradictory information; while some indirectly related or even inaccurate content may help LLM generate more accurate answers by providing suggestive information or logical clues. To solve this, we propose GainRAG, a novel approach that aligns the retriever's and LLM's preferences by defining a new metric, "gain", which measure how well an input passage contributes to correct outputs. Specifically, we propose a method to estimate these gain signals and train a middleware that aligns the preferences of the retriever and the LLM using only limited data. In addition, we introduce a pseudo-passage strategy to mitigate degradation. The experimental results on 6 datasets verify the effectiveness of GainRAG.

cs.IR↗

arXiv2025

Unified Continuous Generative Models

Recent advances in continuous generative models, including multi-step approaches like diffusion and flow-matching (typically requiring 8-1000 sampling steps) and few-step methods such as consistency models (typically 1-8 steps), have demonstrated impressive generative performance. However, existing work often treats these approaches as distinct paradigms, resulting in separate training and sampling methodologies. We introduce a unified framework for training, sampling, and analyzing these models. Our implementation, the Unified Continuous Generative Models Trainer and Sampler (UCGM-{T,S}), achieves state-of-the-art (SOTA) performance. For example, on ImageNet 256x256 using a 675M diffusion transformer, UCGM-T trains a multi-step model achieving 1.30 FID in 20 steps and a few-step model reaching 1.42 FID in just 2 steps. Additionally, applying UCGM-S to a pre-trained model (previously 1.26 FID at 250 steps) improves performance to 1.06 FID in only 40 steps. Code is available at: https://github.com/LINs-lab/UCGM.

cs.LG↗

arXiv2025

Recognize Any Regions

Understanding the semantics of individual regions or patches of unconstrained images, such as open-world object detection, remains a critical yet challenging task in computer vision. Building on the success of powerful image-level vision-language (ViL) foundation models like CLIP, recent efforts have sought to harness their capabilities by either training a contrastive model from scratch with an extensive collection of region-label pairs or aligning the outputs of a detection model with image-level representations of region proposals. Despite notable progress, these approaches are plagued by computationally intensive training requirements, susceptibility to data noise, and deficiency in contextual information. To address these limitations, we explore the synergistic potential of off-the-shelf foundation models, leveraging their respective strengths in localization and semantics. We introduce a novel, generic, and efficient architecture, named RegionSpot, designed to integrate position-aware localization knowledge from a localization foundation model (e.g., SAM) with semantic information from a ViL model (e.g., CLIP). To fully exploit pretrained knowledge while minimizing training overhead, we keep both foundation models frozen, focusing optimization efforts solely on a lightweight attention-based knowledge integration module. Extensive experiments in open-world object recognition show that our RegionSpot achieves significant performance gain over prior alternatives, along with substantial computational savings (e.g., training our model with 3 million data in a single day using 8 V100 GPUs). RegionSpot outperforms GLIP-L by 2.9 in mAP on LVIS val set, with an even larger margin of 13.1 AP for more challenging and rare categories, and a 2.5 AP increase on ODinW. Furthermore, it exceeds GroundingDINO-L by 11.0 AP for rare categories on the LVIS minival set.

cs.CV↗

arXiv2025

Complexity-Scalable Near-Optimal Transceiver Design for Massive MIMO-BICM Systems

Future wireless networks are envisioned to employ multiple-input multiple-output (MIMO) transmissions with large array sizes, and therefore, the adoption of complexity-scalable transceiver becomes important. In this paper, we propose a novel complexity-scalable transceiver design for MIMO systems exploiting bit-interleaved coded modulation (termed MIMO-BICM systems). The proposed scheme leverages the channel bidiagonalization decomposition (CBD), based on which an optimization framework for the precoder and post-processor is developed for maximizing the mutual information (MI) with finite-alphabet inputs. Particularly, we unveil that the desired precoder and post-processor behave distinctively with respect to the operating signal-to-noise ratio (SNR), where the equivalent channel condition number (ECCN) serves as an effective indicator for the overall achievable rate performance. Specifically, at low SNRs, diagonal transmission with a large ECCN is advantageous, while at high SNRs, uniform subchannel gains with a small ECCN are preferred. This allows us to further propose a low-complexity generalized parallel CBD design (GP-CBD) based on Givens rotation according to a well-approximated closed-form performance metric on the achievable rates that takes into account the insights from the ECCN. Numerical results validate the superior performance of the proposed scheme in terms of achievable rate and bit error rate (BER), compared to state-of-the-art designs across various modulation and coding schemes (MCSs).

eess.SP↗

arXiv2025

Liquid: Language Models are Scalable and Unified Multi-modal Generators

We present Liquid, an auto-regressive generation paradigm that seamlessly integrates visual comprehension and generation by tokenizing images into discrete codes and learning these code embeddings alongside text tokens within a shared feature space for both vision and language. Unlike previous multimodal large language model (MLLM), Liquid achieves this integration using a single large language model (LLM), eliminating the need for external pretrained visual embeddings such as CLIP. For the first time, Liquid uncovers a scaling law that performance drop unavoidably brought by the unified training of visual and language tasks diminishes as the model size increases. Furthermore, the unified token space enables visual generation and comprehension tasks to mutually enhance each other, effectively removing the typical interference seen in earlier models. We show that existing LLMs can serve as strong foundations for Liquid, saving 100x in training costs while outperforming Chameleon in multimodal capabilities and maintaining language performance comparable to mainstream LLMs like LLAMA2. Liquid also outperforms models like SD v2.1 and SD-XL (FID of 5.47 on MJHQ-30K), excelling in both vision-language and text-only tasks. This work demonstrates that LLMs such as Qwen2.5 and GEMMA2 are powerful multimodal generators, offering a scalable solution for enhancing both vision-language understanding and generation. The code and models will be released at https://github.com/FoundationVision/Liquid.

cs.CV↗

arXiv2025

FeGe as a building block for the kagome 1:1, 1:6:6, and 1:3:5 families: hidden d-orbital decoupling of flat band sectors, effective models and interaction Hamiltonians

The electronic structure and interactions of kagome materials, such as the 1:1 (FeGe) and 1:6:6 (MgFe6Ge6) classes, are complicated and involve many orbitals and bands around the Fermi level. Current theoretical models treat the systems in an $s$-orbital kagome representation, unsuited and incorrect both quantitatively and qualitatively to the material realities. In this work, we lay the basis of a faithful framework of the electronic model for this large class of materials. We show that the complicated ``spaghetti" of electronic bands near the Fermi level can be decomposed into three groups of Fe d orbitals coupled to specific Ge orbitals via symmetry and chemical analysis. Such a decomposition allows for a clear analytical understanding (leading to different results than the simple s-orbital kagome models) of the flat bands in the system based on the S-matrix formalism of generalized bipartite lattices. Our three minimal Hamiltonians can reproduce the quasiflat bands, van Hove singularities, topology, and Dirac points close to the Fermi level, which we prove by extensive ab initio studies. We also obtain the interacting Hamiltonian for the d orbitals in FeGe using the constraint random phase approximation (cRPA) method, which faithfully describes the antiferromagnetic phase. We then use FeGe as a fundamental LEGO-like building block for a large family of 1:6:6 kagome materials, which can be obtained by doubling and perturbing the FeGe Hamiltonian. We apply the model to its kagome siblings FeSn and CoSn, and also MgFe6Ge6. We further extend the formalism developed for the 1:1 family to the 1:3:5 family AB3Z5 (A = K, Rb, Cs; B = Cr, V, Ti; Z = Sb, Bi), demonstrating the broad applicability of the LEGO-like building block approach. Our work serves as the first complete framework for the study of the interacting phase diagram of kagome compounds.

cond-mat.str-el↗

arXiv2025

Arm DynamIQ Shared Unit and Real-Time: An Empirical Evaluation

The increasing complexity of embedded hardware platforms poses significant challenges for real-time workloads. Architectural features such as Intel RDT, Arm QoS, and Arm MPAM are either unavailable on commercial embedded platforms or designed primarily for server environments optimized for average-case performance and might fail to deliver the expected real-time guarantees. Arm DynamIQ Shared Unit (DSU) includes isolation features-among others, hardware per-way cache partitioning-that can improve the real-time guarantees of complex embedded multicore systems and facilitate real-time analysis. However, the DSU also targets average cases, and its real-time capabilities have not yet been evaluated. This paper presents the first comprehensive analysis of three real-world deployments of the Arm DSU on Rockchip RK3568, Rockchip RK3588, and NVIDIA Orin platforms. We integrate support for the DSU at the operating system and hypervisor level and conduct a large-scale evaluation using both synthetic and real-world benchmarks with varying types and intensities of interference. Our results make extensive use of performance counters and indicate that, although effective, the quality of partitioning and isolation provided by the DSU depends on the type and the intensity of the interfering workloads. In addition, we uncover and analyze in detail the correlation between benchmarks and different types and intensities of interference.

cs.PF↗

arXiv2025

Theory of Superconductivity in LaRu$_3$Si$_2$ and Predictions of New Kagome Flat Band Superconductors

We present a comprehensive investigation of the flat-band kagome superconductor LaRu$_3$Si$_2$, which has recently been reported to host charge density wave (CDW) order above room temperature ($T_{CDW} \simeq 400$ K). The stable crystal structure above the CDW transition is identified via soft phonon condensation and confirmed to be harmonically stable through ab initio calculations, consistent with recent X-ray diffraction refinements. The electron-phonon coupling (EPC) in LaRu$_3$Si$_2$ is found to be mode-selective, primarily driven by strong interactions between Ru-$B_{3u}$ phonons (local $x$-direction, pointing toward the hexagon center) and Ru-$A_g$ electrons (local $d_{x^2-y^2}$ orbital) within the kagome lattice. Using a spring-ball model, we identify this mode-selective EPC as a universal feature of kagome materials. Employing the newly developed Gaussian approximation of the hopping parameters, we derive an analytical expression for the EPC and demonstrate that superconductivity in LaRu$_3$Si$_2$ is mostly driven by the coupling between the kagome $B_{3u}$ phonons and the $A_g$ electrons. The impact of doping is also investigated, revealing that light hole doping (approximately one hole per unit cell) significantly enhances the superconducting critical temperature $T_c$ by 50%, whereas heavy doping induces structural instability and ferromagnetism. Furthermore, high-throughput screening identifies 3063 stable 1:3:2 kagome materials, of which 428 are predicted to exhibit superconductivity with $T_c > 1$ K, and the highest $T_c$ reaching 15 K. These findings establish LaRu$_3$Si$_2$ and related materials as promising platforms for exploring the interplay among kagome flat bands, EPC, and superconductivity. Additionally, they may offer valuable insights into potential limitations on the $T_c$ of flat-band superconductivity in real materials.

cond-mat.supr-con↗

arXiv2025

SA-Occ: Satellite-Assisted 3D Occupancy Prediction in Real World

Existing vision-based 3D occupancy prediction methods are inherently limited in accuracy due to their exclusive reliance on street-view imagery, neglecting the potential benefits of incorporating satellite views. We propose SA-Occ, the first Satellite-Assisted 3D occupancy prediction model, which leverages GPS & IMU to integrate historical yet readily available satellite imagery into real-time applications, effectively mitigating limitations of ego-vehicle perceptions, involving occlusions and degraded performance in distant regions. To address the core challenges of cross-view perception, we propose: 1) Dynamic-Decoupling Fusion, which resolves inconsistencies in dynamic regions caused by the temporal asynchrony between satellite and street views; 2) 3D-Proj Guidance, a module that enhances 3D feature extraction from inherently 2D satellite imagery; and 3) Uniform Sampling Alignment, which aligns the sampling density between street and satellite views. Evaluated on Occ3D-nuScenes, SA-Occ achieves state-of-the-art performance, especially among single-frame methods, with a 39.05% mIoU (a 6.97% improvement), while incurring only 6.93 ms of additional latency per frame. Our code and newly curated dataset are available at https://github.com/chenchen235/SA-Occ.

cs.CV↗

arXiv2025

Robust Super-Moiré in Large Angle Single-Twist Bilayers

Forming long wavelength moiré superlattices (MSL) at small-angle twist van der Waals (vdW) bilayers has been a key approach to creating moiré flat bands. The small-angle twist, however, leads to strong lattice reconstruction, causing domain walls and moiré disorders, which pose considerable challenges in engineering such platforms. At large twist angles, the rigid lattices render a more robust, but shorter wavelength MSL, making it difficult to engineer flat bands. Here, we depict a novel approach to tailoring robust super-moiré (SM) structures that combines the advantages of both small-twist and large-twist transition metal dichalcogenides (TMDs) bilayers using only a single twist angle near a commensurate angle. Structurally, we unveil the spontaneous formation of a periodic arrangement of three inequivalent commensurate moiré (CM) stacking, where the angle deviation from the commensurate angle can tune the periodicity. Electronically, we reveal a large set of van Hove singularities (VHSs) that indicate strong band hybridization, leading to flat bands near the valence band maximum. Our study paves the way for a new platform of robust SM bilayers with structural rigidity and controllable wavelength, extending the investigation of the interplay among band topology, quantum geometry, and moiré superconductivity to the large twist angle regime.

cond-mat.mtrl-sci↗

arXiv2025

Goku: Flow Based Video Generative Foundation Models

This paper introduces Goku, a state-of-the-art family of joint image-and-video generation models leveraging rectified flow Transformers to achieve industry-leading performance. We detail the foundational elements enabling high-quality visual generation, including the data curation pipeline, model architecture design, flow formulation, and advanced infrastructure for efficient and robust large-scale training. The Goku models demonstrate superior performance in both qualitative and quantitative evaluations, setting new benchmarks across major tasks. Specifically, Goku achieves 0.76 on GenEval and 83.65 on DPG-Bench for text-to-image generation, and 84.85 on VBench for text-to-video tasks. We believe that this work provides valuable insights and practical advancements for the research community in developing joint image-and-video generation models.

cs.CV↗

arXiv2025

MapEval: Towards Unified, Robust and Efficient SLAM Map Evaluation Framework

Evaluating massive-scale point cloud maps in Simultaneous Localization and Mapping (SLAM) remains challenging, primarily due to the absence of unified, robust and efficient evaluation frameworks. We present MapEval, an open-source framework for comprehensive quality assessment of point cloud maps, specifically addressing SLAM scenarios where ground truth map is inherently sparse compared to the mapped environment. Through systematic analysis of existing evaluation metrics in SLAM applications, we identify their fundamental limitations and establish clear guidelines for consistent map quality assessment. Building upon these insights, we propose a novel Gaussian-approximated Wasserstein distance in voxelized space, enabling two complementary metrics under the same error standard: Voxelized Average Wasserstein Distance (AWD) for global geometric accuracy and Spatial Consistency Score (SCS) for local consistency evaluation. This theoretical foundation leads to significant improvements in both robustness against noise and computational efficiency compared to conventional metrics. Extensive experiments on both simulated and real-world datasets demonstrate that MapEval achieves at least \SI{100}{}-\SI{500}{} times faster while maintaining evaluation integrity. The MapEval library\footnote{\texttt{https://github.com/JokerJohn/Cloud\_Map\_Evaluation}} will be publicly available to promote standardized map evaluation practices in the robotics community.

cs.RO↗