arXiv subjects

Mingzhe Li

Publications and source records attributed to Mingzhe Li.

At least 37 records · Page 2Linked to original sources

arXiv2026

Distinguishing impurity-induced bound states from Majorana-like zero-energy peaks in strained CsCa2Fe4As4F2 by scanning tunneling microscopy

Iron-based superconductors offer a versatile platform for exploring topological superconductivity and Majorana zero modes (MZMs), with experimental confirmations in Fe(Te,Se), (Li,Fe)OHFeSe and CaKFe4As4 at ambient pressure, as well as in LiFeAs under local strain. The related properties in other iron-based superconductors still need to be explored, especially under the application of local strain. In this study, we conduct scanning tunneling microscopy/spectroscopy measurements on CsCa2Fe4As4F2 crystals under unidirectional local strain. A fully developed superconducting gap with multiple pairs of coherence peaks are observed, and the gap sizes can be significantly modulated by local strain. Spectroscopic measurements on various types of defects including the nonmagnetic Cs-site vacancies consistently reveal pair-breaking effects. These phenomena support a fully gapped multiband superconductivity scenario with sign-changing. Notably, a sharp zero-energy conductance peak (ZECP) is universally observed on a particular type of defects by using a metallic tip, resembling the MZMs observed at interstitial Fe atoms in Fe(Te,Se) [Nat. Phys. 11, 543 (2015)]. However, by using a superconducting tip to enhance energy resolution as well as by studying the ZECP evolution as functions of magnetic field and tunneling transmissivity, we demonstrate that the ZECP originates from nearly degenerate Yu-Shiba-Rusinov states rather than MZMs. Our study not only provides more insights into the superconducting pairing symmetry of CsCa2Fe4As4F2, but also establishes systematic experimental methods for identifying weak impurity state signals and discerning the physical origins of ZECPs.

cond-mat.supr-con↗

arXiv2026

Scanning tunneling microscopy study of helimagnetic monolayer CrBr2 on s-wave superconductor NbSe2: a topologically trivial system due to weak interfacial coupling

Hybrid magnet-superconductor heterostructures attract significant interest for their potential to host unconventional superconductivity, topological superconductivity, and Majorana physics. Transition metal dihalides (MX2, M = transition metal, X = Cl, Br, I) are compelling magnetic candidates due to their novel magnetic structures and possible ferroelectricity. Here, we employ low-temperature scanning tunneling microscopy/spectroscopy to investigate the interfaces fabricated by growing helimagnet candidate CrBr2 on s-wave superconductor NbSe2. Our results reveal that the monolayer CrBr2 is insulating, the measured low-energy electronic states on it derive entirely from the NbSe2 substrate. The superconducting properties of CrBr2/NbSe2 are nearly identical to the bare NbSe2, manifested by the superconducting gap spectra and their temperature and magnetic field dependence, as well as the spatial distribution and bound states of magnetic vortices. Furthermore, in-gap excitations appear only at the dirty edges of CrBr2 islands and are absent from clean edges, suggesting the lack of intrinsic edge states. Taken together, these findings establish the topologically trivial nature of the helimagnetic insulator/s-wave superconductor system CrBr2/NbSe2, attributable to the absence of interfacial superconducting proximity and weak magnetic coupling.

cond-mat.supr-con↗

arXiv2026

Towards Motion Turing Test: Evaluating Human-Likeness in Humanoid Robots

Humanoid robots have achieved significant progress in motion generation and control, exhibiting movements that appear increasingly natural and human-like. Inspired by the Turing Test, we propose the Motion Turing Test, a framework that evaluates whether human observers can discriminate between humanoid robot and human poses using only kinematic information. To facilitate this evaluation, we present the Human-Humanoid Motion (HHMotion) dataset, which consists of 1,000 motion sequences spanning 15 action categories, performed by 11 humanoid models and 10 human subjects. All motion sequences are converted into SMPL-X representations to eliminate the influence of visual appearance. We recruited 30 annotators to rate the human-likeness of each pose on a 0-5 scale, resulting in over 500 hours of annotation. Analysis of the collected data reveals that humanoid motions still exhibit noticeable deviations from human movements, particularly in dynamic actions such as jumping, boxing, and running. Building on HHMotion, we formulate a human-likeness evaluation task that aims to automatically predict human-likeness scores from motion data. Despite recent progress in multimodal large language models, we find that they remain inadequate for assessing motion human-likeness. To address this, we propose a simple baseline model and demonstrate that it outperforms several contemporary LLM-based methods. The dataset, code, and benchmark will be publicly released to support future research in the community.

cs.CV↗

arXiv2026

EDITOR: Effective and Interpretable Prompt Inversion for Text-to-Image Diffusion Models

Text-to-image generation models~(e.g., Stable Diffusion) have achieved significant advancements, enabling the creation of high-quality and realistic images based on textual descriptions. Prompt inversion, the task of identifying the textual prompt used to generate a specific artifact, holds significant potential for applications including data attribution, model provenance, and watermarking validation. Recent studies introduced a delayed projection scheme to optimize for prompts representative of the vocabulary space, though challenges in semantic fluency and efficiency remain. Advanced image captioning models or visual large language models can generate highly interpretable prompts, but they often lack in image similarity. In this paper, we propose a prompt inversion technique called \sys for text-to-image diffusion models, which includes initializing embeddings using a pre-trained image captioning model, refining them through reverse-engineering in the latent space, and converting them to texts using an embedding-to-text model. Our experiments on the widely-used datasets, such as MS COCO, LAION, Flickr and DiffusionDB, show that our method outperforms existing methods in terms of image similarity, textual alignment, prompt interpretability and generalizability. We further illustrate the application of our generated prompts in tasks such as cross-concept image synthesis, concept manipulation, evolutionary multi-concept generation and unsupervised segmentation.

cs.CV↗

arXiv2026

Self-Foveate: Enhancing Diversity and Difficulty of Synthesized Instructions from Unsupervised Text via Multi-Level Foveation

Synthesizing high-quality instruction data from unsupervised text is a promising paradigm for training large language models (LLMs), yet automated methods for this task still exhibit significant limitations in the diversity and difficulty of synthesized instructions. To address these challenges, we propose Self-Foveate, an LLM-driven method for instruction synthesis. Inspired by hierarchical human visual perception, Self-Foveate introduces a "Micro-Scatter-Macro" multi-level foveation methodology that guides the extraction of textual information at three complementary granularities, from fine-grained details through cross-region connections to holistic patterns, thereby enhancing both the diversity and difficulty of synthesized instructions. Furthermore, a re-synthesis module is incorporated to improve the fidelity of instructions to source text and their overall quality. Comprehensive experiments across multiple unsupervised corpora and diverse model architectures demonstrate that Self-Foveate consistently outperforms existing methods. We publicly release our code at https://github.com/Mubuky/Self-Foveate

cs.AI↗

arXiv2026

A multi-mesh adaptive finite element method for solving the Gross-Pitaevskii equation

It is found that the wave functions of the Gross-Pitaevskii equation (GPE) often vary significantly in different spatial regions, with some components exhibiting sharp variations while others remain smooth. Solving the GPE on a single mesh, even with adaptive refinement, can lead to excessive computational costs due to the need to accommodate the most oscillatory solution. To address this issue, we present a multi-mesh adaptive finite element method for solving the GPE. To this end, we first convert it into a time-dependent equation through the imaginary time propagation method. Then the equation is discretized by the backward Euler method temporally and the multi-mesh adaptive finite element method spatially. The proposed method is compared with the single-mesh adaptive method through a series of numerical experiments, which demonstrate that the multi-mesh adaptive method can achieve the same numerical accuracy with less computational consumption.

math.NA↗

arXiv2026

STAR-S: Improving Safety Alignment through Self-Taught Reasoning on Safety Rules

Defending against jailbreak attacks is crucial for the safe deployment of Large Language Models (LLMs). Recent research has attempted to improve safety by training models to reason over safety rules before responding. However, a key issue lies in determining what form of safety reasoning effectively defends against jailbreak attacks, which is difficult to explicitly design or directly obtain. To address this, we propose \textbf{STAR-S} (\textbf{S}elf-\textbf{TA}ught \textbf{R}easoning based on \textbf{S}afety rules), a framework that integrates the learning of safety rule reasoning into a self-taught loop. The core of STAR-S involves eliciting reasoning and reflection guided by safety rules, then leveraging fine-tuning to enhance safety reasoning. Repeating this process creates a synergistic cycle. Improvements in the model's reasoning and interpretation of safety rules allow it to produce better reasoning data under safety rule prompts, which is then utilized for further training. Experiments show that STAR-S effectively defends against jailbreak attacks, outperforming baselines. Code is available at: https://github.com/pikepokenew/STAR_S.git.

cs.AI↗

arXiv2026

Spatiotemporal Detection and Uncertainty Visualization of Atmospheric Blocking Events

Atmospheric blocking events are quasi-stationary high-pressure systems that disrupt the typical paths of polar and subtropical air currents, often producing prolonged extreme weather events such as summer heat waves or winter cold spells. Despite their critical role in shaping mid-latitude weather, accurately modeling and analyzing blocking events in long meteorological records remains a significant challenge. To address this challenge, we present an uncertainty visualization framework for detecting and characterizing atmospheric blocking events. First, we introduce a geometry-based detection and tracking method, evaluated on both pre-industrial climate model simulations (UKESM) and reanalysis data (ERA5), which represent historical Earth observations assimilated from satellite and station measurements onto regular numerical grids using weather models. Second, we propose a suite of uncertainty-aware summaries: contour boxplots that capture representative boundaries and their variability, frequency heatmaps that encode occurrences, and 3D temporal stacks that situate these patterns in time. Third, we demonstrate our framework in a case study of the 2003 European heatwave, mapping the spatiotemporal occurrences of blocking events using these summaries. Collectively, these uncertainty visualizations reveal where blocking events are most likely to occur and how their spatial footprints evolve over time. We envision our framework as a valuable tool for climate scientists and meteorologists: by analyzing how blocking frequency, duration, and intensity vary across regions and climate scenarios, it supports both the study of historical blocking events and the assessment of scenario-dependent climate risks associated with changes in extreme weather linked to blocking.

cs.GR↗

arXiv2025

StableShard: Stable and Scalable Blockchain Sharding with High Concurrency via Collaborative Committees

Sharding enhances blockchain scalability by partitioning nodes into multiple groups for concurrent transaction processing. Configuring a large number of small shards usually helps improve transaction concurrency, but it also increases the fraction of malicious nodes in each shard, easily causing shard corruption and jeopardizing system security. Existing works attempt to improve concurrency by reducing shard sizes while maintaining security, but typically rely on time-consuming recovery of corrupted shards to restore liveness and network-wide consensus. This causes severe system stagnation and limits scalability. To address this, we present StableShard, a sharded blockchain that securely provides high concurrency with stable and scalable performance. The core idea is to carefully co-design the division of labor between proposer shards (PSs) and finalizer committees (FCs): we deliberately assign 1) asymmetric roles and 2) matching parameters to PSs and FCs. Small PSs focus on fast transaction proposal and local validity, while large FCs focus on resolving forks, finalizing PS blocks, and maintaining liveness for faulty PSs via a cross-layer view-change protocol. Moreover, by fine-tuning key system parameters (e.g., shard size, quorum size), we ensure each PS to tolerate <1/2 fraction of malicious nodes without lossing liveness, and allow multiple FCs to securely coexist (each with <1/3 fraction of malicious nodes) for better scalability. Consequently, StableShard can safely configure many smaller PSs to boost concurrency, while FCs and PSs jointly guarantee safety and liveness without system stagnation, leading to stable and scalable performance. Evaluations show that StableShard achieves up to 10x higher throughput than existing solutions and significantly more stable concurrency under attacks.

cs.DC↗

arXiv2025

SpiralShard: Highly Concurrent and Secure Blockchain Sharding via Linked Cross-shard Endorsement

Blockchain sharding improves the scalability of blockchain systems by partitioning the whole blockchain state, nodes, and transaction workloads into different shards. However, existing blockchain sharding systems generally suffer from a small number of shards, resulting in limited concurrency. The main reason is that existing sharding systems require large shard sizes to ensure security. To enhance the concurrency of blockchain sharding securely, we propose SpiralShard. The intuition is to allow the existence of some shards with a larger fraction of malicious nodes (i.e., corrupted shards), thus reducing shard sizes. SpiralShard can configure more and smaller shards for higher concurrency at the same network size. To ensure security with the existence of corrupted shards, we propose the Linked Cross-shard Endorsement (LCE) protocol. According to our LCE protocol, the blocks of each shard are sequentially verified and endorsed (via intra-shard consensus) by a group of shards before being finalized. As a result, a corrupted shard can eliminate forks with the help of the other shards. We implement SpiralShard based on Harmony and conduct extensive evaluations. Experimental results show that, compared with Harmony, SpiralShard achieves around 19x throughput gain under a large network size with 4,000+ nodes.

cs.CR↗

arXiv2025

EmbedGenius: Towards Automated Software Development for Generic Embedded IoT Systems

Embedded IoT system development is crucial for enabling seamless connectivity and functionality across a wide range of applications. However, such a complex process requires cross-domain knowledge of hardware and software and hence often necessitates direct developer involvement, making it labor-intensive, time-consuming, and error-prone. To address this challenge, this paper introduces EmbedGenius, the first fully automated software development platform for general-purpose embedded IoT systems. The key idea is to leverage the reasoning ability of Large Language Models (LLMs) and embedded system expertise to automate the hardware-in-the-loop development process. The main methods include a component-aware library resolution method for addressing hardware dependencies, a library knowledge generation method that injects utility domain knowledge into LLMs, and an auto-programming method that ensures successful deployment. We evaluate EmbedGenius's performance across 71 modules and four mainstream embedded development platforms with over 350 IoT tasks. Experimental results show that EmbedGenius can generate codes with an accuracy of 95.7% and complete tasks with a success rate of 86.5%, surpassing human-in-the-loop baselines by 15.6%--37.7% and 25.5%--53.4%, respectively. We also show EmbedGenius's potential through case studies in environmental monitoring and remote control systems development.

cs.SE↗

arXiv2025

PROMPTMINER: Black-Box Prompt Stealing against Text-to-Image Generative Models via Reinforcement Learning and Fuzz Optimization

Text-to-image (T2I) generative models such as Stable Diffusion and FLUX can synthesize realistic, high-quality images directly from textual prompts. The resulting image quality depends critically on well-crafted prompts that specify both subjects and stylistic modifiers, which have become valuable digital assets. However, the rising value and ubiquity of high-quality prompts expose them to security and intellectual-property risks. One key threat is the prompt stealing attack, i.e., the task of recovering the textual prompt that generated a given image. Prompt stealing enables unauthorized extraction and reuse of carefully engineered prompts, yet it can also support beneficial applications such as data attribution, model provenance analysis, and watermarking validation. Existing approaches often assume white-box gradient access, require large-scale labeled datasets for supervised training, or rely solely on captioning without explicit optimization, limiting their practicality and adaptability. To address these challenges, we propose PROMPTMINER, a black-box prompt stealing framework that decouples the task into two phases: (1) a reinforcement learning-based optimization phase to reconstruct the primary subject, and (2) a fuzzing-driven search phase to recover stylistic modifiers. Experiments across multiple datasets and diffusion backbones demonstrate that PROMPTMINER achieves superior results, with CLIP similarity up to 0.958 and textual alignment with SBERT up to 0.751, surpassing all baselines. Even when applied to in-the-wild images with unknown generators, it outperforms the strongest baseline by 7.5 percent in CLIP similarity, demonstrating better generalization. Finally, PROMPTMINER maintains strong performance under defensive perturbations, highlighting remarkable robustness. Code: https://github.com/aaFrostnova/PromptMiner

cs.CV↗

arXiv2025

Cosmological Perturbations of Extended NGR Model with Parity Violation

Recently the modified teleparallel gravity models attracted a lot of interests. In this paper we consider more extensions to the New General Relativity (NGR) model with parity violations. This extended NGR model differs from the normal NGR model by the inclusion of additional parity-odd term that is quadratic in the torsion tensor. By investigating its cosmological perturbations of this model, we find that this model can avoid ghost instabilities in certain regions of the parameter space, where the coefficient of the parity-odd term does not vanish.

gr-qc↗

arXiv2025

Atomic Smart Contract Interoperability with High Efficiency via Cross-Chain Integrated Execution

With the development of Ethereum, numerous blockchains compatible with Ethereum's execution environment (i.e., Ethereum Virtual Machine, EVM) have emerged. Developers can leverage smart contracts to run various complex decentralized applications on top of blockchains. However, the increasing number of EVM-compatible blockchains has introduced significant challenges in cross-chain interoperability, particularly in ensuring efficiency and atomicity for the whole cross-chain application. Existing solutions are either limited in guaranteeing overall atomicity for the cross-chain application, or inefficient due to the need for multiple rounds of cross-chain smart contract execution. To address this gap, we propose IntegrateX, an efficient cross-chain interoperability system that ensures the overall atomicity of cross-chain smart contract invocations. The core idea is to deploy the logic required for cross-chain execution onto a single blockchain, where it can be executed in an integrated manner. This allows cross-chain applications to perform all cross-chain logic efficiently within the same blockchain. IntegrateX consists of a cross-chain smart contract deployment protocol and a cross-chain smart contract integrated execution protocol. The former achieves efficient and secure cross-chain deployment by decoupling smart contract logic from state, and employing an off-chain cross-chain deployment mechanism combined with on-chain cross-chain verification. The latter ensures atomicity of cross-chain invocations through a 2PC-based mechanism, and enhances performance through transaction aggregation and fine-grained state lock. We implement a prototype of IntegrateX. Extensive experiments demonstrate that it reduces up to 61.2% latency compared to the state-of-the-art baseline while maintaining low gas consumption.

cs.DC↗

arXiv2025

Injecting Domain-Specific Knowledge into Large Language Models: A Comprehensive Survey

Large Language Models (LLMs) have demonstrated remarkable success in various tasks such as natural language understanding, text summarization, and machine translation. However, their general-purpose nature often limits their effectiveness in domain-specific applications that require specialized knowledge, such as healthcare, chemistry, or legal analysis. To address this, researchers have explored diverse methods to enhance LLMs by integrating domain-specific knowledge. In this survey, we provide a comprehensive overview of these methods, which we categorize into four key approaches: dynamic knowledge injection, static knowledge embedding, modular adapters, and prompt optimization. Each approach offers unique mechanisms to equip LLMs with domain expertise, balancing trade-offs between flexibility, scalability, and efficiency. We discuss how these methods enable LLMs to tackle specialized tasks, compare their advantages and disadvantages, evaluate domain-specific LLMs against general LLMs, and highlight the challenges and opportunities in this emerging field. For those interested in delving deeper into this area, we also summarize the commonly used datasets and benchmarks. To keep researchers updated on the latest studies, we maintain an open-source at: https://github.com/abilliyb/Knowledge_Injection_Survey_Papers, dedicated to documenting research in the field of specialized LLM.

cs.CL↗

arXiv2025

Flexible and Probabilistic Topology Tracking with Partial Optimal Transport

In this paper, we present a flexible and probabilistic framework for tracking topological features in time-varying scalar fields using merge trees and partial optimal transport. Merge trees are topological descriptors that record the evolution of connected components in the sublevel sets of scalar fields. We present a new technique for modeling and comparing merge trees using tools from partial optimal transport. In particular, we model a merge tree as a measure network, that is, a network equipped with a probability distribution, and define a notion of distance on the space of merge trees inspired by partial optimal transport. Such a distance offers a new and flexible perspective for encoding intrinsic and extrinsic information in the comparative measures of merge trees. More importantly, it gives rise to a partial matching between topological features in time-varying data, thus enabling flexible topology tracking for scientific simulations. Furthermore, such partial matching may be interpreted as probabilistic coupling between features at adjacent time steps, which gives rise to probabilistic tracking graphs. We derive a stability result for our distance and provide numerous experiments indicating the efficacy of our framework in extracting meaningful feature tracks.

cs.CG↗

arXiv2025

Extremely Scalable Distributed Computation of Contour Trees via Pre-Simplification

Contour trees offer an abstract representation of the level set topology in scalar fields and are widely used in topological data analysis and visualization. However, applying contour trees to large-scale scientific datasets remains challenging due to scalability limitations. Recent developments in distributed hierarchical contour trees have addressed these challenges by enabling scalable computation across distributed systems. Building on these structures, advanced analytical tasks -- such as volumetric branch decomposition and contour extraction -- have been introduced to facilitate large-scale scientific analysis. Despite these advancements, such analytical tasks substantially increase memory usage, which hampers scalability. In this paper, we propose a pre-simplification strategy to significantly reduce the memory overhead associated with analytical tasks on distributed hierarchical contour trees. We demonstrate enhanced scalability through strong scaling experiments, constructing the largest known contour tree -- comprising over half a trillion nodes with complex topology -- in under 15 minutes on a dataset containing 550 billion elements.

cs.CG↗

arXiv2025

SP-Chain: Boosting Intra-Shard and Cross-Shard Security and Performance in Blockchain Sharding

A promising way to overcome the scalability limitations of the current blockchain is to use sharding, which is to split the transaction processing among multiple, smaller groups of nodes. A well-performed blockchain sharding system requires both high performance and high security in both intra- and cross-shard perspectives. However, existing protocols either have issues on protecting security or trade off great performance for security. In this paper, we propose SP-Chain, a blockchain sharding system with enhanced Security and Performance for both intra- and cross-shard perspectives. For intra-shard aspect, we design a two-phase concurrent voting scheme to provide high system throughput and low transaction confirmation latency. Moreover, we propose an efficient unbiased leader rotation scheme to ensure high performance under malicious behavior. For cross-shard aspect, a proof-assisted efficient cross-shard transaction processing mechanism is proposed to guard the cross-shard transactions with low overhead. We implement SP-Chain based on Harmony, and evaluate its performance via large-scale deployment. Extensive evaluations suggest that SP-Chain can process more than 10,000 tx/sec under malicious behaviors with a confirmation latency of 7.6s in a network of 4,000 nodes.

cs.DC↗