arXiv subjects

Mingzhe Li

Publications and source records attributed to Mingzhe Li.

At least 55 records · Page 3Linked to original sources

arXiv2025

Spacetime perturbations and quasi-teleparallel gravity

Gravity is identical to curved spacetime. It is manifested by the curvature of a Riemannian spacetime in general relativity but by torsion or non-metricity in teleparallel gravity models. In this paper, we apply these multiple options to the spacetime perturbation theory and seek the possibilities of representing the gravitation of the background and that of the perturbation in separate ways. We show that the perturbation around a Riemannian background can be described by torsion or non-metricity, so that we have teleparallel like actions for the perturbation.

gr-qc↗

arXiv2025

Flipping Knowledge Distillation: Leveraging Small Models' Expertise to Enhance LLMs in Text Matching

Knowledge distillation typically involves transferring knowledge from a Large Language Model (LLM) to a Smaller Language Model (SLM). However, in tasks such as text matching, fine-tuned smaller models often yield more effective domain-specific representations, as they focus on optimizing the similarity of input pairs. To leverage both the specialized strengths of small models and the rich semantic understanding of LLMs, we introduce a flipped knowledge distillation paradigm, where LLM learns from SLM. Specifically, we address the architectural gap between decoder-only LLMs and smaller encoder-based models by reinterpreting LLMs in an encoder-decoder manner using LoRA. The encoder generates compressed representations, while the decoder maps them to the output space. During training, the encoder produces representations and their similarities, which are then aligned with the similarity scores produced by the teacher, using our proposed Margin-aware Contrastive Learning (MCL) approach. The MCL ensures accurate similarity for both positive and negative pairs, and adaptively handles the internal differences within positive and negative samples. Our paradigm requires only a reasonably good-performing SLM, allowing the LLM to achieve improved performance. Experiments on financial and healthcare benchmarks, as well as real-world applications, confirm its effectiveness, and the model has been fully deployed in an online environment.

cs.CL↗

arXiv2025

Data-Efficient Machine Learning Potentials via Difference Vectors Based on Local Atomic Environments

Constructing efficient and diverse datasets is essential for the development of accurate machine learning potentials (MLPs) in atomistic simulations. However, existing approaches often suffer from data redundancy and high computational costs. Herein, we propose a new method--Difference Vectors based on Local Atomic Environments (DV-LAE)--that encodes structural differences via histogram-based descriptors and enables visual analysis through t-SNE dimensionality reduction. This approach facilitates redundancy detection and dataset optimization while preserving structural diversity. We demonstrate that DV-LAE significantly reduces dataset size and training time across various materials systems, including high-pressure hydrogen, iron-hydrogen binaries, magnesium hydrides, and carbon allotropes, with minimal compromise in prediction accuracy. For instance, in the $α$-Fe/H system, maintaining a highly similar MLP accuracy, the dataset size was reduced by 56%, and the training time per iteration dropped by over 50%. Moreover, we show how visualizing the DV-LAE representation aids in identifying out-of-distribution data by examining the spatial distribution of high-error prediction points, providing a robust reliability metric for new structures during simulations. Our results highlight the utility of local environment visualization not only as an interpretability tool but also as a practical means for accelerating MLP development and ensuring data efficiency in large-scale atomistic modeling.

cs.LG↗

arXiv2025

Demonstration Selection for In-Context Learning via Reinforcement Learning

Diversity in demonstration selection is critical for enhancing model generalization by enabling broader coverage of structures and concepts. Constructing appropriate demonstration sets remains a key research challenge. This paper introduces the Relevance-Diversity Enhanced Selection (RDES), an innovative approach that leverages reinforcement learning (RL) frameworks to optimize the selection of diverse reference demonstrations for tasks amenable to in-context learning (ICL), particularly text classification and reasoning, in few-shot prompting scenarios. RDES employs frameworks like Q-learning and a PPO-based variant to dynamically identify demonstrations that maximize both diversity (quantified by label distribution) and relevance to the task objective. This strategy ensures a balanced representation of reference data, leading to improved accuracy and generalization. Through extensive experiments on multiple benchmark datasets, including diverse reasoning tasks, and involving 14 closed-source and open-source LLMs, we demonstrate that RDES significantly enhances performance compared to ten established baselines. Our evaluation includes analysis of performance across varying numbers of demonstrations on selected datasets. Furthermore, we investigate incorporating Chain-of-Thought (CoT) reasoning, which further boosts predictive performance. The results highlight the potential of RL for adaptive demonstration selection and addressing challenges in ICL.

cs.AI↗

arXiv2025

ManipLVM-R1: Reinforcement Learning for Reasoning in Embodied Manipulation with Large Vision-Language Models

Large Vision-Language Models (LVLMs) have recently advanced robotic manipulation by leveraging vision for scene perception and language for instruction following. However, existing methods rely heavily on costly human-annotated training datasets, which limits their generalization and causes them to struggle in out-of-domain (OOD) scenarios, reducing real-world adaptability. To address these challenges, we propose ManipLVM-R1, a novel reinforcement learning framework that replaces traditional supervision with Reinforcement Learning using Verifiable Rewards (RLVR). By directly optimizing for task-aligned outcomes, our method enhances generalization and physical reasoning while removing the dependence on costly annotations. Specifically, we design two rule-based reward functions targeting key robotic manipulation subtasks: an Affordance Perception Reward to enhance localization of interaction regions, and a Trajectory Match Reward to ensure the physical plausibility of action paths. These rewards provide immediate feedback and impose spatial-logical constraints, encouraging the model to go beyond shallow pattern matching and instead learn deeper, more systematic reasoning about physical interactions.

cs.RO↗

arXiv2025

Yin-Yang vortex on UTe2 (011) surface

UTe2 is a promising candidate for spin-triplet superconductor, yet its exact superconducting order parameter remains highly debated. Here, via scanning tunneling microscopy/spectroscopy, we observe a novel type of magnetic vortex with distinct dark-bright contrast in local density of states on UTe2 (011) surface under a perpendicular magnetic field, resembling the conjugate structure of Yin-Yang diagram in Taoism. Each Yin-Yang vortex contains a quantized magnetic flux, and the boundary between the Yin and Yang parts aligns with the crystallographic a-axis of UTe2. The vortex states exhibit intriguing behaviors -- a sharp zero-energy conductance peak exists at the Yang part, while a superconducting gap with pronounced coherence peaks exists at the Yin part, which is even sharper than those measured far from the vortex core or in the absence of magnetic field. By theoretical modeling, we show that the Yin-Yang vortices on UTe2 (011) surface can be explained by the asymmetric vortex-derived local distortion of the zero-energy surface states associated with spin-triplet pairing with appropriate d-vectors. Therefore, the observation of Yin-Yang vortex confirms the spin-triplet pairing in UTe2 and imposes constraints on the candidate d-vector for the spin-triplet pairing.

cond-mat.supr-con↗

arXiv2025

Audio Jailbreak: An Open Comprehensive Benchmark for Jailbreaking Large Audio-Language Models

The rise of Large Audio Language Models (LAMs) brings both potential and risks, as their audio outputs may contain harmful or unethical content. However, current research lacks a systematic, quantitative evaluation of LAM safety especially against jailbreak attacks, which are challenging due to the temporal and semantic nature of speech. To bridge this gap, we introduce AJailBench, the first benchmark specifically designed to evaluate jailbreak vulnerabilities in LAMs. We begin by constructing AJailBench-Base, a dataset of 1,495 adversarial audio prompts spanning 10 policy-violating categories, converted from textual jailbreak attacks using realistic text to speech synthesis. Using this dataset, we evaluate several state-of-the-art LAMs and reveal that none exhibit consistent robustness across attacks. To further strengthen jailbreak testing and simulate more realistic attack conditions, we propose a method to generate dynamic adversarial variants. Our Audio Perturbation Toolkit (APT) applies targeted distortions across time, frequency, and amplitude domains. To preserve the original jailbreak intent, we enforce a semantic consistency constraint and employ Bayesian optimization to efficiently search for perturbations that are both subtle and highly effective. This results in AJailBench-APT, an extended dataset of optimized adversarial audio samples. Our findings demonstrate that even small, semantically preserved perturbations can significantly reduce the safety performance of leading LAMs, underscoring the need for more robust and semantically aware defense mechanisms.

cs.SD↗

arXiv2025

Tracking Low-Level Cloud Systems with Topology

Low-level clouds are ubiquitous in Earth's atmosphere, playing a crucial role in transporting heat, moisture, and momentum across the planet. Their evolution and interaction with other atmospheric components, such as aerosols, are essential to understanding the climate system and its sensitivity to anthropogenic influences. Advanced high-resolution geostationary satellites now resolve cloud systems with greater accuracy, establishing cloud tracking as a vital research area for studying their spatiotemporal dynamics. It enables disentangling advective and convective components driving cloud evolution. This, in turn, provides deeper insights into the structure and lifecycle of low-level cloud systems and the atmospheric processes they govern. In this paper, we propose a novel framework for tracking cloud systems using topology-driven techniques based on optimal transport. We first obtain a set of anchor points for the cloud systems based on the merge tree of the cloud optical depth field. We then apply topology-driven probabilistic feature tracking of these anchor points to guide the tracking of cloud systems. We demonstrate the utility of our framework by tracking clouds over the ocean and land to test for systematic differences in the two physically distinct settings. We further evaluate our framework through case studies and statistical analyses, comparing it against two leading cloud tracking tools and two topology-based general-purpose tracking tools. The results demonstrate that incorporating system-based tracking improves the ability to capture the evolution of low-level clouds. Our framework paves the way for detailed low-level cloud characterization studies using satellite data records.

cs.CG↗

arXiv2025

Forecasts on Anisotropic Cosmic Birefringence Constraints for CMB Experiment in the Northern Hemisphere

The study of cosmic birefringence through Cosmic Microwave Background (CMB) experiments is a key research area in cosmology and particle physics, providing a critical test for Lorentz and CPT symmetries. This paper focuses on an upcoming CMB experiment in the mid-latitude of the Northern Hemisphere, and investigates the potential to detect anisotropies in cosmic birefringence. Applying a quadratic estimator on simulated polarization data, we reconstruct the power spectrum of anisotropic cosmic birefringence successfully and estimate constraints on the amplitude of the spectrum, $A_{\mathrm{CB}}$, assuming scale invariance. The forecast is based on a wide-scan observation strategy during winter, yielding an effective sky coverage of approximately 23.6%. We consider two noise scenarios corresponding to the short-term and long-term phases of the experiment. Our results show that with a small aperture telescope operating at 95/150GHz, the $2σ$ upper bound for $A_{\mathrm{CB}}$ can reach 0.017 under the low noise scenario when adopting the method of merging multi-frequency data in map domain, and merging multi-frequency data in spectrum domain tightens the limit by about 10%.A large-aperture telescope with the same bands is found to be more effective, tightening the $2σ$ upper limit to 0.0062.

astro-ph.CO↗

arXiv2025

Keywords and Instances: A Hierarchical Contrastive Learning Framework Unifying Hybrid Granularities for Text Generation

Contrastive learning has achieved impressive success in generation tasks to militate the "exposure bias" problem and discriminatively exploit the different quality of references. Existing works mostly focus on contrastive learning on the instance-level without discriminating the contribution of each word, while keywords are the gist of the text and dominant the constrained mapping relationships. Hence, in this work, we propose a hierarchical contrastive learning mechanism, which can unify hybrid granularities semantic meaning in the input text. Concretely, we first propose a keyword graph via contrastive correlations of positive-negative pairs to iteratively polish the keyword representations. Then, we construct intra-contrasts within instance-level and keyword-level, where we assume words are sampled nodes from a sentence distribution. Finally, to bridge the gap between independent contrast levels and tackle the common contrast vanishing problem, we propose an inter-contrast mechanism that measures the discrepancy between contrastive keyword nodes respectively to the instance distribution. Experiments demonstrate that our model outperforms competitive baselines on paraphrasing, dialogue generation, and storytelling tasks.

cs.CL↗

arXiv2025

PAPILLON: Efficient and Stealthy Fuzz Testing-Powered Jailbreaks for LLMs

Large Language Models (LLMs) have excelled in various tasks but are still vulnerable to jailbreaking attacks, where attackers create jailbreak prompts to mislead the model to produce harmful or offensive content. Current jailbreak methods either rely heavily on manually crafted templates, which pose challenges in scalability and adaptability, or struggle to generate semantically coherent prompts, making them easy to detect. Additionally, most existing approaches involve lengthy prompts, leading to higher query costs. In this paper, to remedy these challenges, we introduce a novel jailbreaking attack framework called PAPILLON, which is an automated, black-box jailbreaking attack framework that adapts the black-box fuzz testing approach with a series of customized designs. Instead of relying on manually crafted templates,PAPILLON starts with an empty seed pool, removing the need to search for any related jailbreaking templates. We also develop three novel question-dependent mutation strategies using an LLM helper to generate prompts that maintain semantic coherence while significantly reducing their length. Additionally, we implement a two-level judge module to accurately detect genuine successful jailbreaks. We evaluated PAPILLON on 7 representative LLMs and compared it with 5 state-of-the-art jailbreaking attack strategies. For proprietary LLM APIs, such as GPT-3.5 turbo, GPT-4, and Gemini-Pro, PAPILLONs achieves attack success rates of over 90%, 80%, and 74%, respectively, exceeding existing baselines by more than 60\%. Additionally, PAPILLON can maintain high semantic coherence while significantly reducing the length of jailbreak prompts. When targeting GPT-4, PAPILLON can achieve over 78% attack success rate even with 100 tokens. Moreover, PAPILLON demonstrates transferability and is robust to state-of-the-art defenses. Code: https://github.com/aaFrostnova/Papillon

cs.CR↗

arXiv2025

Pastiche Novel Generation Creating: Fan Fiction You Love in Your Favorite Author's Style

Great novels create immersive worlds with rich character arcs, well-structured plots, and nuanced writing styles. However, current novel generation methods often rely on brief, simplistic story outlines and generate details using plain, generic language. To bridge this gap, we introduce the task of Pastiche Novel Generation, which requires the generated novels to imitate the distinctive features of the original work, including understanding character profiles, predicting plausible plot developments, and writing concrete details using vivid, expressive language. To achieve this, we propose WriterAgent, a novel generation system designed to master the core aspects of literary pastiche. WriterAgent is trained through a curriculum learning paradigm, progressing from low-level stylistic mastery to high-level narrative coherence. Its key tasks include language style learning, character modeling, plot planning, and stylish writing, ensuring comprehensive narrative control. To support this, WriterAgent leverages the WriterLoRA framework, an extension of LoRA with hierarchical and cumulative task-specific modules, each specializing in a different narrative aspect. We evaluate WriterAgent on multilingual classics like Harry Potter and Dream of the Red Chamber, demonstrating its superiority over baselines in capturing the target author's settings, character dynamics, and writing style to produce coherent, faithful narratives.

cs.CL↗

arXiv2025

Direct Visualization of an Incommensurate Unidirectional Charge Density Wave in La$_4$Ni$_3$O$_{10}$

Superconductivity emerges in both La$_3$Ni$_2$O$_7$ and La$_4$Ni$_3$O$_{10}$ under high pressure by suppressing their density-wave transitions, but critical temperature (Tc) differs significantly between these two compounds. To gain deeper insights into the distinct superconducting states, it is essential to unravel the nature of the density-wave states at ambient pressure, a topic that remains largely unexplored. Here, using scanning tunneling microscopy/spectroscopy (STM/STS), we report the direct visualization of an incommensurate unidirectional charge density wave (CDW) in La$_4$Ni$_3$O$_{10}$ in real space. The density of states (DOS) is strongly depleted near $E_F$, indicating the opening of a CDW gap of $2Δ \approx 71$ meV, which is unfavorable for the formation of superconductivity at ambient pressure. We propose that the CDW arises from Fermi surface nesting, and is likely a subsidiary phase of a spin density wave. Compared to La$_3$Ni$_2$O$_7$, the weaker electronic correlation in La$_4$Ni$_3$O$_{10}$ is likely one reason for the lower $T_c$.

cond-mat.supr-con↗

arXiv2025

Universal Catalyst Design Framework for Electrochemical Hydrogen Peroxide Synthesis Facilitated by Local Atomic Environment Descriptors

Developing a universal and precise design framework is crucial to search high-performance catalysts, but it remains a giant challenge due to the diverse structures and sites across various types of catalysts. To address this challenge, herein, we developed a novel framework by the refined local atomic environment descriptors (i.e., weighted Atomic Center Symmetry Function, wACSF) combined with machine learning (ML), microkinetic modeling, and computational high-throughput screening. This framework is successfully integrated into the Digital Catalysis Database (DigCat), enabling efficient screening for 2e- water oxidation reaction (2e- WOR) catalysts across four material categories (i.e., metal alloys, metal oxides and perovskites, and single-atom catalysts) within a ML model. The proposed wACSF descriptors integrating both geometric and chemical features are proven effective in predicting the adsorption free energies with ML. Excitingly, based on the wACSF descriptors, the ML models accurately predict the adsorption free energies of hydroxyl (ΔGOH*) and oxygen (ΔGO*) for such a wide range of catalysts, achieving R2 values of 0.84 and 0.91, respectively. Through density functional theory calculations and microkinetic modeling, a universal 2e- WOR microkinetic volcano model was derived with excellent agreement with experimental observations reported to date, which was further used to rapidly screen high-performance catalysts with the input of ML-predicted ΔGOH*. Most importantly, this universal framework can significantly improve the efficiency of catalyst design by considering multiple types of materials at the same time, which can dramatically accelerate the screening of high-performance catalysts.

cond-mat.mtrl-sci↗

arXiv2024

Blockchain Data Analysis in the Era of Large-Language Models

Blockchain data analysis is essential for deriving insights, tracking transactions, identifying patterns, and ensuring the integrity and security of decentralized networks. It plays a key role in various areas, such as fraud detection, regulatory compliance, smart contract auditing, and decentralized finance (DeFi) risk management. However, existing blockchain data analysis tools face challenges, including data scarcity, the lack of generalizability, and the lack of reasoning capability. We believe large language models (LLMs) can mitigate these challenges; however, we have not seen papers discussing LLM integration in blockchain data analysis in a comprehensive and systematic way. This paper systematically explores potential techniques and design patterns in LLM-integrated blockchain data analysis. We also outline prospective research opportunities and challenges, emphasizing the need for further exploration in this promising field. This paper aims to benefit a diverse audience spanning academia, industry, and policy-making, offering valuable insights into the integration of LLMs in blockchain data analysis.

cs.CR↗

arXiv2024

Instability of the charge density wave in Kagome magnet FeGe

Kagome metals show rich competing quantum phases due to geometry frustration, flat bands, many-body effects, and non-trivial topology. Recently, a novel charge density wave (CDW) was discovered deep inside the antiferromagnetic phase of FeGe, attracting intense attention due to close relation with magnetism. Here, via a scanning tunneling microscope (STM), we find the 2*2 CDW in FeGe is very fragile and can be readily disrupted into the initial 1*1 phase; Small sqrt3*sqrt3 CDW puddles are found to coexist with the 2*2 CDW in as-grown samples, and can also be induced in the intermediate process of CDW disruption, which will eventually transform into the initial 1*1 phase. Moreover, an exotic intermediate CDW state and standalone CDW nuclei appear unexpectedly during the disruption process. Our first-principle calculations find equal softening of a flat optical phonon mode in a large momentum region around the CDW wave vector, corresponding to numerous competing CDWs with close energies. This might lead to strong instability of the CDW ground state, responsible for STM observations. Our findings provide more novel experimental aspects to understand the CDW in FeGe and suggest FeGe-like Kagome metals are ideal platforms for studying the physics of competing CDW instabilities.

cond-mat.str-el↗

arXiv2024

Observation of Yu-Shiba-Rusinov-like states at the edge of CrBr3/NbSe2 heterostructure

The hybrid ferromagnet-superconductor heterostructures have attracted extensive attention as they potentially host topological superconductivity. Relevant experimental signatures have recently been reported in CrBr3/NbSe2 ferromagnet-superconductor heterostructure, but controversies remain. Here, we reinvestigate CrBr3/NbSe2 by an ultralow temperature scanning tunneling microscope with higher spatial and energy resolutions. We find that the single-layer CrBr3 film is insulating and acts likely as a vacuum barrier, the measured superconducting gap and vortex state on it are nearly the same as those of NbSe2 substrate. Meanwhile, in-gap features are observed at the edges of CrBr3 island, which display either a zero-energy conductance peak or a pair of particle-hole symmetric bound states. They are discretely distributed at the edges of CrBr3 film, and their appearance is found closely related to the atomic lattice reconstruction near the edges. By increasing tunneling transmissivity, the zero-energy conductance peak quickly splits, while the pair of nonzero in-gap bound states first approach each other, merge, and then split again. These behaviors are unexpected for Majorana edge modes, but in consistent with the conventional Yu-Shiba-Rusinov states. Our results provide critical information for further understanding the interfacial coupling in CrBr3/NbSe2 heterostructure.

cond-mat.supr-con↗

arXiv2024

MStableChain: Towards Multi-Native Stablecoins in EVM-Compatible Blockchain for Stable Fee and Mass Adoption

Traditional blockchain systems, such as Ethereum, typically rely on a \emph{single volatile cryptocurrency for transaction fees}. This leads to fluctuating transaction fee prices and limits the flexibility of users' payment options. To address these issues, we propose MStableChain, which leverage multiple stablecoins as native tokens for transaction fee settlements, thus ensuring stable transaction fees and flexible payment options. To address the challenges of mass adoption and practicality, we propose several core designs. To maintain compatibility with the Ethereum Virtual Machine (EVM) for mass adoption while supporting multiple native stablecoins, MStableChain employs a multi-currency units, multi-type RPCs mechanism. This mechanism enables the system to handle multiple stablecoins without altering the EVM or requiring changes to user applications. Furthermore, an oracle-based gas fee adjustment mechanism is proposed to manage exchange rates between different stablecoins, ensuring equitable transaction costs across various currencies. The system also introduces a secure, on-chain voting-based management protocol for the administrative functions related to these stablecoins. Experimental results from a prototype implementation demonstrate that MStableChain provides stable transaction fee prices, high effectiveness, and good usability.

cs.CE↗