arXiv subjects

Rui Zhang

Publications and source records attributed to Rui Zhang.

At least 415 records · Page 23Linked to original sources

arXiv2025

BO-SA-PINNs: Self-adaptive physics-informed neural networks based on Bayesian optimization for automatically designing PDE solvers

Physics-informed neural networks (PINNs) is becoming a popular alternative method for solving partial differential equations (PDEs). However, they require dedicated manual modifications to the hyperparameters of the network, the sampling methods and loss function weights for different PDEs, which reduces the efficiency of the solvers. In this paper, we pro- pose a general multi-stage framework, i.e. BO-SA-PINNs to alleviate this issue. In the first stage, Bayesian optimization (BO) is used to select hyperparameters for the training process, and based on the results of the pre-training, the network architecture, learning rate, sampling points distribution and loss function weights suitable for the PDEs are automatically determined. The proposed hyperparameters search space based on experimental results can enhance the efficiency of BO in identifying optimal hyperparameters. After selecting the appropriate hyperparameters, we incorporate a global self-adaptive (SA) mechanism the second stage. Using the pre-trained model and loss information in the second-stage training, the exponential moving average (EMA) method is employed to optimize the loss function weights, and residual-based adaptive refinement with distribution (RAR-D) is used to optimize the sampling points distribution. In the third stage, L-BFGS is used for stable training. In addition, we introduce a new activation function that enables BO-SA-PINNs to achieve higher accuracy. In numerical experiments, we conduct comparative and ablation experiments to verify the performance of the model on Helmholtz, Maxwell, Burgers and high-dimensional Poisson equations. The comparative experiment results show that our model can achieve higher accuracy and fewer iterations in test cases, and the ablation experiments demonstrate the positive impact of every improvement.

cs.CE↗

arXiv2025

Unlearnable Examples Give a False Sense of Data Privacy: Understanding and Relearning

Unlearnable examples are proposed to prevent third parties from exploiting unauthorized data, which generates unlearnable examples by adding imperceptible perturbations to public publishing data. These unlearnable examples proficiently misdirect the model training process, leading it to focus on learning perturbation features while neglecting the semantic features of the image. In this paper, we make an in-depth analysis and observe that models can learn both image features and perturbation features of unlearnable examples at an early training stage, but are rapidly trapped in perturbation features learning since the shallow layers tend to learn on perturbation features and propagate harmful activations to deeper layers. Based on the observations, we propose Progressive Staged Training, a self-adaptive training framework specially designed to break unlearnable examples. The proposed framework effectively prevents models from becoming trapped in learning perturbation features. We evaluated our method on multiple model architectures over diverse datasets, e.g., CIFAR-10, CIFAR-100, and ImageNet-mini. Our method circumvents the unlearnability of all state-of-the-art methods in the literature, revealing that existing unlearnable examples give a false sense of privacy protection and provide a reliable baseline for further evaluation of unlearnable techniques.

cs.CV↗

arXiv2025

Exploring the Impact of Personality Traits on Conversational Recommender Systems: A Simulation with Large Language Models

Conversational Recommender Systems (CRSs) engage users in multi-turn interactions to deliver personalized recommendations. The emergence of large language models (LLMs) further enhances these systems by enabling more natural and dynamic user interactions. However, a key challenge remains in understanding how personality traits shape conversational recommendation outcomes. Psychological evidence highlights the influence of personality traits on user interaction behaviors. To address this, we introduce an LLM-based personality-aware user simulation for CRSs (PerCRS). The user agent induces customizable personality traits and preferences, while the system agent possesses the persuasion capability to simulate realistic interaction in CRSs. We incorporate multi-aspect evaluation to ensure robustness and conduct extensive analysis from both user and system perspectives. Experimental results demonstrate that state-of-the-art LLMs can effectively generate diverse user responses aligned with specified personality traits, thereby prompting CRSs to dynamically adjust their recommendation strategies. Our experimental analysis offers empirical insights into the impact of personality traits on the outcomes of conversational recommender systems.

cs.CL↗

arXiv2025

MACER3D -- an upgrade of MACER2D with enhanced subgrid models and gas physics -- and its application to simulating AGN feedback in a massive elliptical galaxy

We present MACER3D (Multiscale AGN-regulated Cosmic Ecosystem Resolver in 3D), a new suite of three-dimensional hydrodynamic simulations that study active galactic nuclei (AGN) feedback on galactic scales over Gyr duration, with major enhancement in subgrid models and gas physics over its predecessor -- MACER (Massive AGN Controlled Ellipticals Resolved) which is in two dimensions (hereafter MACER2D). MACER3D resolves gas dynamics from within the Bondi radius ($\sim 25\,\mathrm{pc}$) to halo scales. Combined with black hole accretion theory, it enables an accurate calculation of AGN outputs and subsequently their large-scale feedback effects. We present results from simulating an isolated elliptical galaxy with different feedback configurations. In the fiducial model with both AGN and supernova (SN) feedback, the temporal evolution of AGN luminosity and star formation rate are strongly correlated, suggesting shared dependence on the availability of gas supply for SMBH accretion and star formation. AGN duty cycles of several percent with a single-cycle timescale of $\sim 10^2\,\mathrm{Myr}$ agree with observations, while models with only AGN or SN feedback fail to reproduce observed cycles. While all models maintain a quiescent galaxy state, fiducial AGN+SN feedback model results in higher star formation than no-SN feedback, suggesting SN feedback, when acting synergistically with AGN feedback, may positively impact star formation. Combined AGN and SN feedback enhances halo-scale metal enrichment compared to single-feedback models. The simulated X-ray properties match observations and predict transient cavities produced by cold-mode AGN winds from past burst events. The differences between the results obtained by MACER2D and MACER3D are also discussed.

astro-ph.GA↗

arXiv2025

SiReRAG: Indexing Similar and Related Information for Multihop Reasoning

Indexing is an important step towards strong performance in retrieval-augmented generation (RAG) systems. However, existing methods organize data based on either semantic similarity (similarity) or related information (relatedness), but do not cover both perspectives comprehensively. Our analysis reveals that modeling only one perspective results in insufficient knowledge synthesis, leading to suboptimal performance on complex tasks requiring multihop reasoning. In this paper, we propose SiReRAG, a novel RAG indexing approach that explicitly considers both similar and related information. On the similarity side, we follow existing work and explore some variances to construct a similarity tree based on recursive summarization. On the relatedness side, SiReRAG extracts propositions and entities from texts, groups propositions via shared entities, and generates recursive summaries to construct a relatedness tree. We index and flatten both similarity and relatedness trees into a unified retrieval pool. Our experiments demonstrate that SiReRAG consistently outperforms state-of-the-art indexing methods on three multihop datasets (MuSiQue, 2WikiMultiHopQA, and HotpotQA), with an average 1.9% improvement in F1 scores. As a reasonably efficient solution, SiReRAG enhances existing reranking methods significantly, with up to 7.8% improvement in average F1 scores. Our code is available at https://github.com/SalesforceAIResearch/SiReRAG .

cs.CL↗

arXiv2025

GReaTer: Gradients over Reasoning Makes Smaller Language Models Strong Prompt Optimizers

The effectiveness of large language models (LLMs) is closely tied to the design of prompts, making prompt optimization essential for enhancing their performance across a wide range of tasks. Many existing approaches to automating prompt engineering rely exclusively on textual feedback, refining prompts based solely on inference errors identified by large, computationally expensive LLMs. Unfortunately, smaller models struggle to generate high-quality feedback, resulting in complete dependence on large LLM judgment. Moreover, these methods fail to leverage more direct and finer-grained information, such as gradients, due to operating purely in text space. To this end, we introduce GReaTer, a novel prompt optimization technique that directly incorporates gradient information over task-specific reasoning. By utilizing task loss gradients, GReaTer enables self-optimization of prompts for open-source, lightweight language models without the need for costly closed-source LLMs. This allows high-performance prompt optimization without dependence on massive LLMs, closing the gap between smaller models and the sophisticated reasoning often needed for prompt refinement. Extensive evaluations across diverse reasoning tasks including BBH, GSM8k, and FOLIO demonstrate that GReaTer consistently outperforms previous state-of-the-art prompt optimization methods, even those reliant on powerful LLMs. Additionally, GReaTer-optimized prompts frequently exhibit better transferability and, in some cases, boost task performance to levels comparable to or surpassing those achieved by larger language models, highlighting the effectiveness of prompt optimization guided by gradients over reasoning. Code of GReaTer is available at https://github.com/psunlpgroup/GreaTer.

cs.CL↗

arXiv2025

A global adaptive velocity space for general discrete velocity framework in predictions of rarefied and multi-scale flows

The rarefied flow and multi-scale flow are crucial for the aerodynamic design of spacecraft, ultra-low orbital vehicles and plumes. By introducing a discrete velocity space, the Boltzmann method, such as the discrete velocity method and unified methods, can capture complex and non-equilibrium velocity distribution functions (VDFs) and describe flow behaviors exactly. However, the extremely steep slope and high concentration of the gas VDFs in a local particle velocity space make it very difficult for the Boltzmann method with structured velocity space to describe high speed flow. Therefore, the adaptive velocity space (AVS) is required for the Boltzmann solvers to be practical in complex rarefied flow and multi-scale flow. This paper makes two improvements to the AVS approach, which is then incorporated into a general discrete velocity framework, such as the unified gas-kinetic scheme. Firstly, a global velocity mesh is used to prevent the interpolation of the VDFs at the physical interface during the calculation of the microscopic fluxes, maintaining the program's high level of parallelism. Secondly, rather than utilizing costly interpolation, the VDFs on a new velocity space were reconstruction using the ``consanguinity" relationship. In other words, a split child node's VDF is the same as its parent's VDF, and a merged parent's VDF is the average of its children's VDFs. Additionally, the discrete deviation of the equilibrium distribution functions is employed to maintain the proposed method's conservation. Moreover, an appropriate set of adaptive parameters is established to enhance the automation of the proposed method. Finally, a number of numerical tests are carried out to validate the proposed method.

physics.flu-dyn↗

arXiv2025

GREATERPROMPT: A Unified, Customizable, and High-Performing Open-Source Toolkit for Prompt Optimization

LLMs have gained immense popularity among researchers and the general public for its impressive capabilities on a variety of tasks. Notably, the efficacy of LLMs remains significantly dependent on the quality and structure of the input prompts, making prompt design a critical factor for their performance. Recent advancements in automated prompt optimization have introduced diverse techniques that automatically enhance prompts to better align model outputs with user expectations. However, these methods often suffer from the lack of standardization and compatibility across different techniques, limited flexibility in customization, inconsistent performance across model scales, and they often exclusively rely on expensive proprietary LLM APIs. To fill in this gap, we introduce GREATERPROMPT, a novel framework that democratizes prompt optimization by unifying diverse methods under a unified, customizable API while delivering highly effective prompts for different tasks. Our framework flexibly accommodates various model scales by leveraging both text feedback-based optimization for larger LLMs and internal gradient-based optimization for smaller models to achieve powerful and precise prompt improvements. Moreover, we provide a user-friendly Web UI that ensures accessibility for non-expert users, enabling broader adoption and enhanced performance across various user groups and application scenarios. GREATERPROMPT is available at https://github.com/psunlpgroup/GreaterPrompt via GitHub, PyPI, and web user interfaces.

cs.LG↗

arXiv2025

A novel Cercignani-Lampis boundary model for discrete velocity methods in predicting rarefied and multi-scale flows

To extend the discrete velocity method (DVM) and unified methods to more realistic boundary conditions, a Cercignani-Lampis (CL) boundary with different momentum and thermal energy accommodations is proposed and integrated into the DVM framework. By giving the macroscopic flux from the numerical quadrature of the incident molecular distribution, the reflected macroscopic flux can be obtained for the given accommodation coefficients. Then, an anisotropic Gaussian distribution can be found for the reflected molecules, whose parameters are determined by the calculated reflected macroscopic flux. These macroscopic flux and microscopic Gaussian distribution form a complete physical process for the reflected molecules. Furthermore, the CL boundary is integrated into the unified gas-kinetic scheme (UGKS), making it suitable for the simulation of both monatomic and diatomic gas flows, and it accommodates both the conventional Cartesian velocity space and the recently developed efficient unstructured velocity space. Moreover, this new GSI boundary is suitable for both explicit and implicit schemes, offering better performance for flow prediction. Finally, the performance of the new boundary is validated through a series of numerical tests covering a wide range of Knudsen and Mach numbers.

physics.flu-dyn↗

arXiv2025

Joint Active and Passive Beamforming Design for IRS-aided MIMO ISAC Based on Sensing Mutual Information

In this paper, we investigate the intelligent reflecting surface (IRS)/reconfigurable intelligent surface (RIS)-aided integrated sensing and communication (ISAC) system based on sensing mutual information (MI). Specifically, the base station (BS) perceives the sensing target via the reflected sensing signal by the IRS, while communicating with the users simultaneously. Our aim is to maximize the sensing MI, subject to the quality of service (QoS) constraints for all communication users, the transmit power constraint at the BS, and the unit-modulus constraint on the IRS's passive reflection. We solve this problem under two cases: one simplified case assuming a line-of-sight (LoS) channel between the BS and IRS and no clutter interference to sensing, and the other generalized case considering the Rician fading channel of the BS-IRS link and the presence of clutter interference to sensing. For the first case, we prove that the dedicated sensing beamformer is unnecessary for improving sensing MI and develop a low-complexity iterative algorithm to jointly optimize the BS and IRS active/passive beamformers. Then, for the second case, we propose an alternative iterative algorithm, which can also be applied to the first case, to solve the beamforming design problem under the general setup. Numerical results are provided to validate the performance of the proposed algorithms, as compared to various benchmark schemes.

eess.SP↗

arXiv2025

CancerLLM: A Large Language Model in Cancer Domain

Medical Large Language Models (LLMs) have demonstrated impressive performance on a wide variety of medical NLP tasks; however, there still lacks a LLM specifically designed for phenotyping identification and diagnosis in cancer domain. Moreover, these LLMs typically have several billions of parameters, making them computationally expensive for healthcare systems. Thus, in this study, we propose CancerLLM, a model with 7 billion parameters and a Mistral-style architecture, pre-trained on nearly 2.7M clinical notes and over 515K pathology reports covering 17 cancer types, followed by fine-tuning on two cancer-relevant tasks, including cancer phenotypes extraction and cancer diagnosis generation. Our evaluation demonstrated that the CancerLLM achieves state-of-the-art results with F1 score of 91.78% on phenotyping extraction and 86.81% on disganois generation. It outperformed existing LLMs, with an average F1 score improvement of 9.23%. Additionally, the CancerLLM demonstrated its efficiency on time and GPU usage, and robustness comparing with other LLMs. We demonstrated that CancerLLM can potentially provide an effective and robust solution to advance clinical research and practice in cancer domain

cs.CL↗

arXiv2025

Machine Learning Assisted Modeling of Amorphous TiO$_2$-Doped GeO$_2$ for Advanced LIGO Mirror Coatings

The mechanical loss angle of amorphous TiO$_2$-doped GeO$_2$ can be lower than 10$^{-4}$, making it a candidate for Laser Interferometer Gravitational-wave Observatory (LIGO) mirror coatings. Amorphous oxides have complex atomic structures that are influenced by various factors, including doping concentration, preparation, and thermal history, resulting in different mass densities and physical properties. Modeling at atomistic level enables capturing these effects by generating atomic structure models according to experimental conditions. In order to obtain reliable and physical amorphous models at an affordable cost, we develop classical and machine-learning potentials (MLP) to speed up simulations. First-principles calculations are used to train and validate MLP as well as validating structure models. To better reproduce properties such as elastic modulus, radial distribution function (RDF) and the variations in mass density of doped amorphous oxides, density functional theory (DFT) calculations are used to optimize the final models. We find that the mass densities of amorphous systems are correlated with the total void volume. The experimental mass density matches the models with the most symmetric potential energy wells under volume change. The elastic response of the metal-oxygen network is also studied. The 27\% TiO$_2$ doped GeO$_2$ system shows the least number of large atom-atom distance changes, while for 44\% TiO$_2$ doped GeO$_2$, a majority of Ti-O distances are significantly changed. In response to strains, the metal-oxygen network at low mass densities prefers to adjust bond angles, while at high mass densities, the adjustment is mainly done by changing atom-atom distance.

cond-mat.mtrl-sci↗

arXiv2025

OmniDocBench: Benchmarking Diverse PDF Document Parsing with Comprehensive Annotations

Document content extraction is a critical task in computer vision, underpinning the data needs of large language models (LLMs) and retrieval-augmented generation (RAG) systems. Despite recent progress, current document parsing methods have not been fairly and comprehensively evaluated due to the narrow coverage of document types and the simplified, unrealistic evaluation procedures in existing benchmarks. To address these gaps, we introduce OmniDocBench, a novel benchmark featuring high-quality annotations across nine document sources, including academic papers, textbooks, and more challenging cases such as handwritten notes and densely typeset newspapers. OmniDocBench supports flexible, multi-level evaluations--ranging from an end-to-end assessment to the task-specific and attribute--based analysis using 19 layout categories and 15 attribute labels. We conduct a thorough evaluation of both pipeline-based methods and end-to-end vision-language models, revealing their strengths and weaknesses across different document types. OmniDocBench sets a new standard for the fair, diverse, and fine-grained evaluation in document parsing. Dataset and code are available at https://github.com/opendatalab/OmniDocBench.

cs.CV↗

arXiv2025

Coverage-based Fairness in Multi-document Summarization

Fairness in multi-document summarization (MDS) measures whether a system can generate a summary fairly representing information from documents with different social attribute values. Fairness in MDS is crucial since a fair summary can offer readers a comprehensive view. Previous works focus on quantifying summary-level fairness using Proportional Representation, a fairness measure based on Statistical Parity. However, Proportional Representation does not consider redundancy in input documents and overlooks corpus-level unfairness. In this work, we propose a new summary-level fairness measure, Equal Coverage, which is based on coverage of documents with different social attribute values and considers the redundancy within documents. To detect the corpus-level unfairness, we propose a new corpus-level measure, Coverage Parity. Our human evaluations show that our measures align more with our definition of fairness. Using our measures, we evaluate the fairness of thirteen different LLMs. We find that Claude3-sonnet is the fairest among all evaluated LLMs. We also find that almost all LLMs overrepresent different social attribute values. The code is available at https://github.com/leehaoyuan/coverage_fairness.

cs.CL↗

arXiv2025

Image Over Text: Transforming Formula Recognition Evaluation with Character Detection Matching

Formula recognition presents significant challenges due to the complicated structure and varied notation of mathematical expressions. Despite continuous advancements in formula recognition models, the evaluation metrics employed by these models, such as BLEU and Edit Distance, still exhibit notable limitations. They overlook the fact that the same formula has diverse representations and is highly sensitive to the distribution of training data, thereby causing unfairness in formula recognition evaluation. To this end, we propose a Character Detection Matching (CDM) metric, ensuring the evaluation objectivity by designing an image-level rather than a LaTeX-level metric score. Specifically, CDM renders both the model-predicted LaTeX and the ground-truth LaTeX formulas into image-formatted formulas, then employs visual feature extraction and localization techniques for precise character-level matching, incorporating spatial position information. Such a spatially-aware and character-matching method offers a more accurate and equitable evaluation compared with previous BLEU and Edit Distance metrics that rely solely on text-based character matching. Experimentally, we evaluated various formula recognition models using CDM, BLEU, and ExpRate metrics. Their results demonstrate that the CDM aligns more closely with human evaluation standards and provides a fairer comparison across different models by eliminating discrepancies caused by diverse formula representations. Code is available at https://github.com/opendatalab/UniMERNet/tree/main/cdm.

cs.CV↗

arXiv2025

Rotatable Antenna Enabled Wireless Communication System with Visual Recognition: A Prototype Implementation

Rotatable antenna (RA) is an emerging technology that has great potential to exploit additional spatial degrees of freedom (DoFs) by flexibly altering the three-dimensional (3D) orientation/boresight of each antenna. In this demonstration, we present a prototype of the RA-enabled wireless communication system with a visual recognition module to evaluate the performance gains provided by the RA in practical environments. In particular, a mechanically-driven RA is developed by integrating a digital servo motor, a directional antenna, and a microcontroller, which enables the dynamic adjustment of the RA orientation. Moreover, the orientation adjustment of the RA is guided by the user's direction information provided by the visual recognition module, thereby significantly enhancing system response speed and self-orientation accuracy. The experimental results demonstrate that the RA-enabled communication system achieves significant improvement in communication coverage performance compared to the conventional fixed antenna system.

eess.SY↗

arXiv2025

A Language Vision Model Approach for Automated Tumor Contouring in Radiation Oncology

Background: Lung cancer ranks as the leading cause of cancer-related mortality worldwide. The complexity of tumor delineation, crucial for radiation therapy, requires expertise often unavailable in resource-limited settings. Artificial Intelligence(AI), particularly with advancements in deep learning (DL) and natural language processing (NLP), offers potential solutions yet is challenged by high false positive rates. Purpose: The Oncology Contouring Copilot (OCC) system is developed to leverage oncologist expertise for precise tumor contouring using textual descriptions, aiming to increase the efficiency of oncological workflows by combining the strengths of AI with human oversight. Methods: Our OCC system initially identifies nodule candidates from CT scans. Employing Language Vision Models (LVMs) like GPT-4V, OCC then effectively reduces false positives with clinical descriptive texts, merging textual and visual data to automate tumor delineation, designed to elevate the quality of oncology care by incorporating knowledge from experienced domain experts. Results: Deployments of the OCC system resulted in a significant reduction in the false discovery rate by 35.0%, a 72.4% decrease in false positives per scan, and an F1-score of 0.652 across our dataset for unbiased evaluation. Conclusions: OCC represents a significant advance in oncology care, particularly through the use of the latest LVMs to improve contouring results by (1) streamlining oncology treatment workflows by optimizing tumor delineation, reducing manual processes; (2) offering a scalable and intuitive framework to reduce false positives in radiotherapy planning using LVMs; (3) introducing novel medical language vision prompt techniques to minimize LVMs hallucinations with ablation study, and (4) conducting a comparative analysis of LVMs, highlighting their potential in addressing medical language vision challenges.

eess.IV↗

arXiv2025

Safety Aware Task Planning via Large Language Models in Robotics

The integration of large language models (LLMs) into robotic task planning has unlocked better reasoning capabilities for complex, long-horizon workflows. However, ensuring safety in LLM-driven plans remains a critical challenge, as these models often prioritize task completion over risk mitigation. This paper introduces SAFER (Safety-Aware Framework for Execution in Robotics), a multi-LLM framework designed to embed safety awareness into robotic task planning. SAFER employs a Safety Agent that operates alongside the primary task planner, providing safety feedback. Additionally, we introduce LLM-as-a-Judge, a novel metric leveraging LLMs as evaluators to quantify safety violations within generated task plans. Our framework integrates safety feedback at multiple stages of execution, enabling real-time risk assessment, proactive error correction, and transparent safety evaluation. We also integrate a control framework using Control Barrier Functions (CBFs) to ensure safety guarantees within SAFER's task planning. We evaluated SAFER against state-of-the-art LLM planners on complex long-horizon tasks involving heterogeneous robotic agents, demonstrating its effectiveness in reducing safety violations while maintaining task efficiency. We also verify the task planner and safety planner through actual hardware experiments involving multiple robots and a human.

cs.RO↗