arXiv subjects

Pengfei Li

Publications and source records attributed to Pengfei Li.

At least 109 records · Page 6Linked to original sources

arXiv2025

LLM-PySC2: Starcraft II learning environment for Large Language Models

The tremendous potential has been demonstrated by large language models (LLMs) in intelligent decision-making problems, with unprecedented capabilities shown across diverse applications ranging from gaming AI systems to complex strategic planning frameworks. However, the StarCraft II platform, which has been widely adopted for validating decision-making algorithms in the past decade, has not yet provided substantial support for this emerging domain. To address issues that LLMs cannot interface with the hundreds of actions of the pysc2 backend and the lack of native support for multi-agent (MA) collaboration, we propose the LLM-PySC2 environment. This is the first environment that offers LLMs the complete pysc2 action space with sufficient multi-modal information and game Wiki knowledge. With an asynchronous query architecture, the environment efficiently interacts with LLMs that maintain a constant latency regardless of the scale of the agents' population. In the experiments, we evaluated LLMs' decision-making performance in both the macro-decision and micro-operation scenarios, with traditional StarCraft II Multi-Agent Challenge (SMAC) tasks and a series of new proposed. Results indicate that LLMs possess the potential to achieve victories in complex scenarios but cannot constantly generate correct decisions, especially in the recovered pysc2 action space and MA settings. Without task-relevant instructions, the pre-trained models suffer from issues such as hallucinations and inefficient collaboration. Our findings suggest that StarCraft II still challenges in the era of large models, revealing that there is a lot to do to develop an advanced LLM decision-making system, and the proposed LLM-PySC2 environment will support future development of LLM-based decision-making solutions.

cs.AI↗

arXiv2025

Retrieval Augmented Learning: A Retrial-based Large Language Model Self-Supervised Learning and Autonomous Knowledge Generation

The lack of domain-specific data in the pre-training of Large Language Models (LLMs) severely limits LLM-based decision systems in specialized applications, while post-training a model in the scenarios requires significant computational resources. In this paper, we present Retrial-Augmented Learning (RAL), a reward-free self-supervised learning framework for LLMs that operates without model training. By developing Retrieval-Augmented Generation (RAG) into a module for organizing intermediate data, we realized a three-stage autonomous knowledge generation of proposing a hypothesis, validating the hypothesis, and generating the knowledge. The method is evaluated in the LLM-PySC2 environment, a representative decision-making platform that combines sufficient complexity with domain-specific knowledge requirements. Experiments demonstrate that the proposed method effectively reduces hallucination by generating and utilizing validated knowledge, and increases decision-making performance at an extremely low cost. Meanwhile, the approach exhibits potential in out-of-distribution(OOD) tasks, robustness, and transferability, making it a cost-friendly but effective solution for decision-making problems and autonomous knowledge generation.

cs.AI↗

arXiv2025

Slow, Nanometer Light Confinement Observed in Atomically Thin TaS2

Extreme light confinement down to the atomic scale has been theoretically predicted for ultrathin, Ta-based transition metal dichalcogenides (TMDs). In this work, we experimentally demonstrate in 2H-TaS$_2$ monolayers and bilayers a lateral confinement ratio up to 300 at large wave vectors of $q = 0.15 \, Å^{-1}$, and slow light behaviour with a group velocity $\sim 10^{-4}c$. Quantitative momentum-resolved electron energy loss spectroscopy (q-EELS) with a momentum resolution of $0.0056 \, Å^{-1}$ was used as a platform for the nanoscale optical measurements. With it, momentum-dispersed, two-dimensional (2D) plasmon resonances were experimentally observed, showing a transition from 2D to 3D Coulomb interaction in the high-momentum regime, equivalent to light confinement volumes of $1\text{-}2 \, \text{nm}^3$. Remarkably, the resonant modes do not enter the electron-hole continuum, predicting even further enhanced optical field confinements for this material at cryogenic temperatures.

cond-mat.mtrl-sci↗

arXiv2025

DIMT25@ICDAR2025: HW-TSC's End-to-End Document Image Machine Translation System Leveraging Large Vision-Language Model

This paper presents the technical solution proposed by Huawei Translation Service Center (HW-TSC) for the "End-to-End Document Image Machine Translation for Complex Layouts" competition at the 19th International Conference on Document Analysis and Recognition (DIMT25@ICDAR2025). Leveraging state-of-the-art open-source large vision-language model (LVLM), we introduce a training framework that combines multi-task learning with perceptual chain-of-thought to develop a comprehensive end-to-end document translation system. During the inference phase, we apply minimum Bayesian decoding and post-processing strategies to further enhance the system's translation capabilities. Our solution uniquely addresses both OCR-based and OCR-free document image translation tasks within a unified framework. This paper systematically details the training methods, inference strategies, LVLM base models, training data, experimental setups, and results, demonstrating an effective approach to document image machine translation.

cs.CV↗

arXiv2025

Pogorelov type $C^2$ estimates for sum Hessian equations

In this paper, We establish Pogorelov type $C^2$ estimates for the admissible solutions with $σ_k(D^2u)$ bounded from below of Sum Hessian equations. We also proved the lower bounded condition can be removed when $k = n$.

math.AP↗

arXiv2025

MMTL-UniAD: A Unified Framework for Multimodal and Multi-Task Learning in Assistive Driving Perception

Advanced driver assistance systems require a comprehensive understanding of the driver's mental/physical state and traffic context but existing works often neglect the potential benefits of joint learning between these tasks. This paper proposes MMTL-UniAD, a unified multi-modal multi-task learning framework that simultaneously recognizes driver behavior (e.g., looking around, talking), driver emotion (e.g., anxiety, happiness), vehicle behavior (e.g., parking, turning), and traffic context (e.g., traffic jam, traffic smooth). A key challenge is avoiding negative transfer between tasks, which can impair learning performance. To address this, we introduce two key components into the framework: one is the multi-axis region attention network to extract global context-sensitive features, and the other is the dual-branch multimodal embedding to learn multimodal embeddings from both task-shared and task-specific features. The former uses a multi-attention mechanism to extract task-relevant features, mitigating negative transfer caused by task-unrelated features. The latter employs a dual-branch structure to adaptively adjust task-shared and task-specific parameters, enhancing cross-task knowledge transfer while reducing task conflicts. We assess MMTL-UniAD on the AIDE dataset, using a series of ablation studies, and show that it outperforms state-of-the-art methods across all four tasks. The code is available on https://github.com/Wenzhuo-Liu/MMTL-UniAD.

cs.CV↗

arXiv2025

Bench4Merge: A Comprehensive Benchmark for Merging in Realistic Dense Traffic with Micro-Interactive Vehicles

While the capabilities of autonomous driving have advanced rapidly, merging into dense traffic remains a significant challenge, many motion planning methods for this scenario have been proposed but it is hard to evaluate them. Most existing closed-loop simulators rely on rule-based controls for other vehicles, which results in a lack of diversity and randomness, thus failing to accurately assess the motion planning capabilities in highly interactive scenarios. Moreover, traditional evaluation metrics are insufficient for comprehensively evaluating the performance of merging in dense traffic. In response, we proposed a closed-loop evaluation benchmark for assessing motion planning capabilities in merging scenarios. Our approach involves other vehicles trained in large scale datasets with micro-behavioral characteristics that significantly enhance the complexity and diversity. Additionally, we have restructured the evaluation mechanism by leveraging Large Language Models (LLMs) to assess each autonomous vehicle merging onto the main lane. Extensive experiments and test-vehicle deployment have demonstrated the progressiveness of this benchmark. Through this benchmark, we have obtained an evaluation of existing methods and identified common issues. The simulation environment and evaluation process can be accessed at https://github.com/WZM5853/Bench4Merge.

cs.RO↗

arXiv2025

Realization of a non-Hermitian Haldane model in circuits

The Haldane model is the simplest yet most powerful topological lattice model exhibiting various phases, including the Dirac semimetal phase and the anomalous quantum Hall phase (also known as the Chern insulator). Although considered unlikely to be physically directly realizable in condensed matter systems, it has been experimentally demonstrated in other physical settings such as cold atoms, where Hermiticity is usually preserved. Extending this model to the non-Hermitian regime with energy non-conservation can significantly enrich topological phases that lack Hermitian counterparts; however, such exploration remains experimentally challenging due to the lack of suitable physical platforms. Here, based on electric circuits, we report the experimental realization of a genuine non-Hermitian Haldane model with asymmetric next-nearest-neighbor hopping. We observe two previously uncovered phases: a non-Hermitian Chern insulator and a non-Hermitian semimetal phase, both exhibiting boundary-dependent amplifying or dissipative chiral edge states. Our work paves the way for exploring non-Hermiticity-induced unconventional topological phases in the Haldane model.

physics.optics↗

arXiv2025

Making AI Less "Thirsty": Uncovering and Addressing the Secret Water Footprint of AI Models

The growing carbon footprint of artificial intelligence (AI) has been undergoing public scrutiny. Nonetheless, the equally important water (withdrawal and consumption) footprint of AI has largely remained under the radar. For example, training the GPT-3 language model in Microsoft's state-of-the-art U.S. data centers can directly evaporate 700,000 liters of clean freshwater, but such information has been kept a secret. More critically, the global AI demand is projected to account for 4.2-6.6 billion cubic meters of water withdrawal in 2027, which is more than the total annual water withdrawal of 4-6 Denmark or half of the United Kingdom. This is concerning, as freshwater scarcity has become one of the most pressing challenges. To respond to the global water challenges, AI can, and also must, take social responsibility and lead by example by addressing its own water footprint. In this paper, we provide a principled methodology to estimate the water footprint of AI, and also discuss the unique spatial-temporal diversities of AI's runtime water efficiency. Finally, we highlight the necessity of holistically addressing water footprint along with carbon footprint to enable truly sustainable AI.

cs.LG↗

arXiv2025

Strain-engineering quantum anomalous Hall effect in janus MnBi2SexTe4-x monolayers

Exploring intrinsic magnetic topological insulators (TIs) for next-generation spintronic devices is still challenging in recent years. Here, we present a theoretical investigation on the electronic, magnetic and topological properties of monolayer (ML) Janus MnBi2TexSe4-x, derived from two trivial magnetic semiconductors ML MnBi2Se4 and MnBi2Te4. Our band structure analysis reveals that two out of the eight Janus structures exhibit band inversion induced by spin-orbit coupling. These structures are confirmed to have nonzero integer Chern numbers, indicating their topological nature. Moreover, the topological state is robust under moderate biaxial strains. Interestingly, applying compressive strain results in a high Chern number of 2 and enhances their magnetic stability at elevated temperatures. Our findings offer an effective strategy to engineer magnetic TI states within the ML MnBi2Te4 family.

cond-mat.mtrl-sci↗

arXiv2025

Positive and Unlabeled Data: Model, Estimation, Inference, and Classification

This study introduces a new approach to addressing positive and unlabeled (PU) data through the double exponential tilting model (DETM). Traditional methods often fall short because they only apply to selected completely at random (SCAR) PU data, where the labeled positive and unlabeled positive data are assumed to be from the same distribution. In contrast, our DETM's dual structure effectively accommodates the more complex and underexplored selected at random PU data, where the labeled and unlabeled positive data can be from different distributions. We rigorously establish the theoretical foundations of DETM, including identifiability, parameter estimation, and asymptotic properties. Additionally, we move forward to statistical inference by developing a goodness-of-fit test for the SCAR condition and constructing confidence intervals for the proportion of positive instances in the target domain. We leverage an approximated Bayes classifier for classification tasks, demonstrating DETM's robust performance in prediction. Through theoretical insights and practical applications, this study highlights DETM as a comprehensive framework for addressing the challenges of PU data.

stat.ME↗

arXiv2025

Likelihood-based Nonparametric Receiver Operating Characteristic Curve Analysis in the Presence of Imperfect Reference Standard

In diagnostic studies, researchers frequently encounter imperfect reference standards with some misclassified labels. Treating these as gold standards can bias receiver operating characteristic (ROC) curve analysis. To address this issue, we propose a novel likelihood-based method under a nonparametric density ratio model. This approach enables the reliable estimation of the ROC curve, area under the curve (AUC), partial AUC, and Youden's index with favorable statistical properties. To implement the method, we develop an efficient expectation-maximization algorithm algorithm. Extensive simulations evaluate its finite-sample performance, showing smaller mean squared errors in estimating the ROC curve, partial AUC, and Youden's index compared to existing methods. We apply the proposed approach to a malaria study.

stat.ME↗

arXiv2025

Effects of fractional diffraction on nonlinear PT phase transitions and stability of dark solitons and vortices

The wave propagation under the action of fractional diffraction has recently drawn increasing attention in nonlinear optics. Here, we address the effect of fractional diffraction on the existence, phase transitions, and stability of dark solitons (DSs) and vortices in parity-time (PT) symmetric graded-index waveguide with self-defocusing nonlinearity. The DSs and vortices are produced by numerical solution of the corresponding one- and two-dimensional fractional nonlinear Schrödinger equations. We show that solution branches of fundamental and higher-order DSs collide pair-wise (merge) and disappear with the increase of the gain-loss strength, revealing nonlinear PT phase transitions in the waveguide. Numerically identifying the merger points, we demonstrate effects of the fractional diffraction on the phase transition.The phase transition points determine boundaries of existence regions for the DSs and vortices.The stability of the DSs and vortices is studied by means of the linearization with respect to small perturbations. Direct simulations of perturbed evolution corroborate their stability properties predicted by the analysis of small perturbations.

physics.optics↗

arXiv2025

Semi-Supervised Vision-Centric 3D Occupancy World Model for Autonomous Driving

Understanding world dynamics is crucial for planning in autonomous driving. Recent methods attempt to achieve this by learning a 3D occupancy world model that forecasts future surrounding scenes based on current observation. However, 3D occupancy labels are still required to produce promising results. Considering the high annotation cost for 3D outdoor scenes, we propose a semi-supervised vision-centric 3D occupancy world model, PreWorld, to leverage the potential of 2D labels through a novel two-stage training paradigm: the self-supervised pre-training stage and the fully-supervised fine-tuning stage. Specifically, during the pre-training stage, we utilize an attribute projection head to generate different attribute fields of a scene (e.g., RGB, density, semantic), thus enabling temporal supervision from 2D labels via volume rendering techniques. Furthermore, we introduce a simple yet effective state-conditioned forecasting module to recursively forecast future occupancy and ego trajectory in a direct manner. Extensive experiments on the nuScenes dataset validate the effectiveness and scalability of our method, and demonstrate that PreWorld achieves competitive performance across 3D occupancy prediction, 4D occupancy forecasting and motion planning tasks.

cs.CV↗

arXiv2025

Asymptotical Behavior of Global Solutions of the Navier-Stokes-Korteweg Equations with Respect to Capillarity Number at Infinity

Vanishing capillarity in the Navier-Stokes-Korteweg (NSK) equations has been widely investigated, in particular, it is well-known that the NSK equations converge to the Navier-Stokes (NS) equations by vanishing capillarity number. To our best knowledge, this paper first investigates the behavior of large capillary number, denoted by $κ^2$, for the global(-in-time) strong solutions with small initial perturbations of the three-dimensional (3D) NSK equations in a slab domain with Navier(-slip) boundary condition. Under the well-prepared initial data, we can construct a family of global strong solutions of the 3D incompressible NSK equations with respect to $κ>0$, where the solutions converge to a unique solution of 2D incompressible NS-like equations as $κ$ goes to infinity.

math.AP↗

arXiv2025

Learning-Augmented Online Control for Decarbonizing Water Infrastructures

Water infrastructures are essential for drinking water supply, irrigation, fire protection, and other critical applications. However, water pumping systems, which are key to transporting water to the point of use, consume significant amounts of energy and emit millions of tons of greenhouse gases annually. With the wide deployment of digital water meters and sensors in these infrastructures, Machine Learning (ML) has the potential to optimize water supply control and reduce greenhouse gas emissions. Nevertheless, the inherent vulnerability of ML methods in terms of worst-case performance raises safety concerns when deployed in critical water infrastructures. To address this challenge, we propose a learning-augmented online control algorithm, termed LAOC, designed to dynamically schedule the activation and/or speed of water pumps. To ensure safety, we introduce a novel design of safe action sets for online control problems. By leveraging these safe action sets, LAOC can provably guarantee safety constraints while utilizing ML predictions to reduce energy and environmental costs. Our analysis reveals the tradeoff between safety requirements and average energy/environmental cost performance. Additionally, we conduct an experimental study on a building water supply system to demonstrate the empirical performance of LAOC. The results indicate that LAOC can effectively reduce environmental and energy costs while guaranteeing safety constraints.

eess.SY↗

arXiv2025

PeFoMed: Parameter Efficient Fine-tuning of Multimodal Large Language Models for Medical Imaging

Multimodal large language models (MLLMs) represent an evolutionary expansion in the capabilities of traditional large language models, enabling them to tackle challenges that surpass the scope of purely text-based applications. It leverages the knowledge previously encoded within these language models, thereby enhancing their applicability and functionality in the reign of multimodal contexts. Recent works investigate the adaptation of MLLMs as a universal solution to address medical multi-modal problems as a generative task. In this paper, we propose a parameter efficient framework for fine-tuning MLLMs, specifically validated on medical visual question answering (Med-VQA) and medical report generation (MRG) tasks, using public benchmark datasets. We also introduce an evaluation metric using the 5-point Likert scale and its weighted average value to measure the quality of the generated reports for MRG tasks, where the scale ratings are labelled by both humans manually and the GPT-4 model. We further assess the consistency of performance metrics across traditional measures, GPT-4, and human ratings for both VQA and MRG tasks. The results indicate that semantic similarity assessments using GPT-4 align closely with human annotators and provide greater stability, yet they reveal a discrepancy when compared to conventional lexical similarity measurements. This questions the reliability of lexical similarity metrics for evaluating the performance of generative models in Med-VQA and report generation tasks. Besides, our fine-tuned model significantly outperforms GPT-4v. This indicates that without additional fine-tuning, multi-modal models like GPT-4v do not perform effectively on medical imaging tasks. The code will be available here: https://github.com/jinlHe/PeFoMed.

cs.CL↗

arXiv2025

MiniMax-01: Scaling Foundation Models with Lightning Attention

We introduce MiniMax-01 series, including MiniMax-Text-01 and MiniMax-VL-01, which are comparable to top-tier models while offering superior capabilities in processing longer contexts. The core lies in lightning attention and its efficient scaling. To maximize computational capacity, we integrate it with Mixture of Experts (MoE), creating a model with 32 experts and 456 billion total parameters, of which 45.9 billion are activated for each token. We develop an optimized parallel strategy and highly efficient computation-communication overlap techniques for MoE and lightning attention. This approach enables us to conduct efficient training and inference on models with hundreds of billions of parameters across contexts spanning millions of tokens. The context window of MiniMax-Text-01 can reach up to 1 million tokens during training and extrapolate to 4 million tokens during inference at an affordable cost. Our vision-language model, MiniMax-VL-01 is built through continued training with 512 billion vision-language tokens. Experiments on both standard and in-house benchmarks show that our models match the performance of state-of-the-art models like GPT-4o and Claude-3.5-Sonnet while offering 20-32 times longer context window. We publicly release MiniMax-01 at https://github.com/MiniMax-AI.

cs.CL↗