arXiv subjects

Bo Zhang

Publications and source records attributed to Bo Zhang.

At least 307 records · Page 17Linked to original sources

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

Pharmacokinetic characteristics of Jinhong tablets in normal, chronic superficial gastritis and intestinal microbial disorder rats

Jinhong tablet (JHT), a traditional Chinese medicine made from four herbs, effectively treats chronic superficial gastritis (CSG) by soothing the liver, relieving depression, regulating qi, and promoting blood circulation. However, its pharmacokinetics are underexplored. This study investigates JHT's pharmacokinetics in normal rats and its differences in normal, CSG, and intestinal microbial disorder rats. A quantitative method for seven active ingredients in rat plasma was established using UPLC-TQ-MS/MS. After administering various JHT doses, plasma concentrations were measured to assess pharmacokinetics in normal rats. The pharmacokinetics of four main ingredients were compared in normal, CSG, and fecal microbiota transplantation (FMT) rats. Intestinal microbial changes were evaluated by high-throughput sequencing. Spearman correlation analysis linked ingredient exposure to gut microbiota disturbances. The method showed good linearity, precision, accuracy, extraction recovery, and stability. In normal rats, all seven ingredients were rapidly absorbed. Tetrahydropalmatine, corydaline, costunolide, and rhamnosylvitexin had good exposure, while dehydrocorydaline, allocryptopine, and palmatine hydrochloride had low exposure. Tetrahydropalmatine, corydaline, and costunolide followed linear pharmacokinetics (AUC0-t, Cmax) at doses of 0.7-5.6 g/kg, while rhamnosylvitexin and dehydrocorydaline showed linearity at 0.7-2.8 g/kg. In CSG and FMT rats, pharmacokinetic differences were observed. CSG enhanced costunolide exposure and Cmax, and increased rhamnosylvitexin exposure. FMT raised corydaline exposure and rhamnosylvitexin Cmax, linked to 20 bacterial genera.

q-bio.QM↗

arXiv2025

Multi-Dimensional AGV Path Planning in 3D Warehouses Using Ant Colony Optimization and Advanced Neural Networks

Within modern warehouse scenarios, the rapid expansion of e-commerce and increasingly complex, multi-level storage environments have exposed the limitations of traditional AGV (Automated Guided Vehicle) path planning methods--often reliant on static 2D models and expert-tuned heuristics that struggle to handle dynamic traffic and congestion. Addressing these limitations, this paper introduces a novel AGV path planning approach for 3D warehouse environments that leverages a hybrid framework combining ACO (Ant Colony Optimization) with deep learning models, called NAHACO (Neural Adaptive Heuristic Ant Colony Optimization). NAHACO integrates three key innovations: first, an innovative heuristic algorithm for 3D warehouse cargo modeling using multidimensional tensors, which addresses the challenge of achieving superior heuristic accuracy; second, integration of a congestion-aware loss function within the ACO framework to adjust path costs based on traffic and capacity constraints, called CARL (Congestion-Aware Reinforce Loss), enabling dynamic heuristic calibration for optimizing ACO-based path planning; and third, an adaptive attention mechanism that captures multi-scale spatial features, thereby addressing dynamic heuristic calibration for further optimization of ACO-based path planning and AGV navigation. NAHACO significantly boosts path planning efficiency, yielding faster computation times and superior performance over both vanilla and state-of-the-art methods, while automatically adapting to warehouse constraints for real-time optimization. NAHACO outperforms state-of-the-art methods, lowering the total cost by up to 24.7% on TSP benchmarks. In warehouse tests, NAHACO cuts cost by up to 41.5% and congestion by up to 56.1% compared to previous methods.

cs.RO↗

arXiv2025

Identification of a dwarf galaxy stream in Gaia, and its possible association with the VPOS structure

Low surface density streams are important tracers to study the formation and evolution of the Milky Way. Using the accurate astrometric measurements from Gaia mission, we discover a low surface density stream in the north hemisphere, with length of $\sim110$ degree and width of $1.23$ kpc. The vertical velocity dispersion perpendicular to the stream is around $22.4$ km s$^{-1}$. The spectral data of member candidate stars from LAMOST and DESI shows a large metallicity range from $-1.8$ to $-0.7$. Based on those properties we claim that the stream is originated from a dwarf galaxy. The median metallicity of $\mathrm{[Fe/H]}=-1.3$ indicates a massive dwarf galaxy origination with stellar mass around $2.0\times10^7M_\odot$, which is comparable with the Fornax dwarf galaxy and smaller than LMC/SMC and Sagittarius. We also find the globular cluster Pyxis is highly associated with the stream in the phase space $E-L_Z$ and metallicity. The massive progenitor also suggests that many dwarf galaxies, including massive ones, have been disrupted during their evolution orbiting the Milky Way and left with very low surface density structures. This is important to understand the {\it missing satellites} problem. The orbit information of the stream shows tight association between its progenitor and the Vast POlar Structure (VPOS), which indicates that the satellites fell into the Milky Way in groups, which brought many globular clusters into the Milky Way.

astro-ph.GA↗

arXiv2025

LeX-Art: Rethinking Text Generation via Scalable High-Quality Data Synthesis

We introduce LeX-Art, a comprehensive suite for high-quality text-image synthesis that systematically bridges the gap between prompt expressiveness and text rendering fidelity. Our approach follows a data-centric paradigm, constructing a high-quality data synthesis pipeline based on Deepseek-R1 to curate LeX-10K, a dataset of 10K high-resolution, aesthetically refined 1024$\times$1024 images. Beyond dataset construction, we develop LeX-Enhancer, a robust prompt enrichment model, and train two text-to-image models, LeX-FLUX and LeX-Lumina, achieving state-of-the-art text rendering performance. To systematically evaluate visual text generation, we introduce LeX-Bench, a benchmark that assesses fidelity, aesthetics, and alignment, complemented by Pairwise Normalized Edit Distance (PNED), a novel metric for robust text accuracy evaluation. Experiments demonstrate significant improvements, with LeX-Lumina achieving a 79.81% PNED gain on CreateBench, and LeX-FLUX outperforming baselines in color (+3.18%), positional (+4.45%), and font accuracy (+3.81%). Our codes, models, datasets, and demo are publicly available.

cs.CV↗

arXiv2025

Lumina-Image 2.0: A Unified and Efficient Image Generative Framework

We introduce Lumina-Image 2.0, an advanced text-to-image generation framework that achieves significant progress compared to previous work, Lumina-Next. Lumina-Image 2.0 is built upon two key principles: (1) Unification - it adopts a unified architecture (Unified Next-DiT) that treats text and image tokens as a joint sequence, enabling natural cross-modal interactions and allowing seamless task expansion. Besides, since high-quality captioners can provide semantically well-aligned text-image training pairs, we introduce a unified captioning system, Unified Captioner (UniCap), specifically designed for T2I generation tasks. UniCap excels at generating comprehensive and accurate captions, accelerating convergence and enhancing prompt adherence. (2) Efficiency - to improve the efficiency of our proposed model, we develop multi-stage progressive training strategies and introduce inference acceleration techniques without compromising image quality. Extensive evaluations on academic benchmarks and public text-to-image arenas show that Lumina-Image 2.0 delivers strong performances even with only 2.6B parameters, highlighting its scalability and design efficiency. We have released our training details, code, and models at https://github.com/Alpha-VLLM/Lumina-Image-2.0.

cs.CV↗

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

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

R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalization

Large Language Models have demonstrated remarkable reasoning capability in complex textual tasks. However, multimodal reasoning, which requires integrating visual and textual information, remains a significant challenge. Existing visual-language models often struggle to effectively analyze and reason visual content, resulting in suboptimal performance on complex reasoning tasks. Moreover, the absence of comprehensive benchmarks hinders the accurate assessment of multimodal reasoning capabilities. In this paper, we introduce R1-Onevision, a multimodal reasoning model designed to bridge the gap between visual perception and deep reasoning. To achieve this, we propose a cross-modal reasoning pipeline that transforms images into formal textural representations, enabling precise language-based reasoning. Leveraging this pipeline, we construct the R1-Onevision dataset which provides detailed, step-by-step multimodal reasoning annotations across diverse domains. We further develop the R1-Onevision model through supervised fine-tuning and reinforcement learning to cultivate advanced reasoning and robust generalization abilities. To comprehensively evaluate multimodal reasoning performance across different grades, we introduce R1-Onevision-Bench, a benchmark aligned with human educational stages, covering exams from junior high school to university and beyond. Experimental results show that R1-Onevision achieves state-of-the-art performance, outperforming models such as GPT-4o and Qwen2.5-VL on multiple challenging multimodal reasoning benchmarks.

cs.CV↗

arXiv2025

Pulsation Properties of Blazhko and Non-Blazhko RRab Stars

In this study, we conduct a comparative analysis of the properties of Blazhko and non-Blazhko RRab stars. We identified 1054 non-Blazhko and 785 Blazhko RRab stars in the photometric data observed by K2 mission, which, combined with those 37 stars observed in the original Kepler field, constituted our study sample. Using the Fourier Decomposition method, we calculated the pulsation parameters, including phase differences and amplitude ratios, for these RRab stars, revealing significant discrepancies in the pulsation parameters between Blazhko and non-Blazhko RRab stars. However, distinguishing between Blazhko and Non-Blazhko RRab stars based on Fourier parameters remains challenging due to the significant overlap in their distributions. By cross-matching our sample with the LRS of LAMOST DR12, we identified 147 Blazhko and 111 non-Blazhko RRab stars, which exhibit similar metallicity distributions. Furthermore, cross-matching with Gaia DR3 data yielded 766 Blazhko and 950 non-Blazhko RRab stars, showing differences in color indices but not in absolute magnitudes. Our findings suggested the Blazhko effect is linked to pulsation parameters and colors, rather than metallicities or absolute magnitude.

astro-ph.SR↗

arXiv2025

Corrected Riemann smoothed particle hydrodynamics method for multi-resolution fluid-structure interaction

As a mesh-free method, smoothed particle hydrodynamics (SPH) has been widely used for modeling and simulating fluid-structure interaction (FSI) problems. While the kernel gradient correction (KGC) method is commonly applied in structural domains to enhance numerical consistency, high-order consistency corrections that preserve conservation remain underutilized in fluid domains despite their critical role in FSI analysis, especially for the multi-resolution scheme where fluid domains generally have a low resolution. In this study, we incorporate the reverse kernel gradient correction (RKGC) formulation, a conservative high-order consistency approximation, into the fluid discretization for solving FSI problems. RKGC has been proven to achieve exact second-order convergence with relaxed particles and improve numerical accuracy while particularly enhancing energy conservation in free-surface flow simulations. By integrating this correction into the Riemann SPH method to solve different typical FSI problems with a multi-resolution scheme, numerical results consistently show improvements in accuracy and convergence compared to uncorrected fluid discretization. Despite these advances, further refinement of correction techniques for solid domains and fluid-structure interfaces remains significant for enhancing the overall accuracy of SPH-based FSI modeling and simulation.

cs.CE↗

arXiv2025

PerturboLLaVA: Reducing Multimodal Hallucinations with Perturbative Visual Training

This paper aims to address the challenge of hallucinations in Multimodal Large Language Models (MLLMs) particularly for dense image captioning tasks. To tackle the challenge, we identify the current lack of a metric that finely measures the caption quality in concept level. We hereby introduce HalFscore, a novel metric built upon the language graph and is designed to evaluate both the accuracy and completeness of dense captions at a granular level. Additionally, we identify the root cause of hallucination as the model's over-reliance on its language prior. To address this, we propose PerturboLLaVA, which reduces the model's reliance on the language prior by incorporating adversarially perturbed text during training. This method enhances the model's focus on visual inputs, effectively reducing hallucinations and producing accurate, image-grounded descriptions without incurring additional computational overhead. PerturboLLaVA significantly improves the fidelity of generated captions, outperforming existing approaches in handling multimodal hallucinations and achieving improved performance across general multimodal benchmarks.

cs.CV↗

arXiv2025

An End-to-End Learning-Based Multi-Sensor Fusion for Autonomous Vehicle Localization

Multi-sensor fusion is essential for autonomous vehicle localization, as it is capable of integrating data from various sources for enhanced accuracy and reliability. The accuracy of the integrated location and orientation depends on the precision of the uncertainty modeling. Traditional methods of uncertainty modeling typically assume a Gaussian distribution and involve manual heuristic parameter tuning. However, these methods struggle to scale effectively and address long-tail scenarios. To address these challenges, we propose a learning-based method that encodes sensor information using higher-order neural network features, thereby eliminating the need for uncertainty estimation. This method significantly eliminates the need for parameter fine-tuning by developing an end-to-end neural network that is specifically designed for multi-sensor fusion. In our experiments, we demonstrate the effectiveness of our approach in real-world autonomous driving scenarios. Results show that the proposed method outperforms existing multi-sensor fusion methods in terms of both accuracy and robustness. A video of the results can be viewed at https://youtu.be/q4iuobMbjME.

cs.RO↗

arXiv2025

SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing

Survey paper plays a crucial role in scientific research, especially given the rapid growth of research publications. Recently, researchers have begun using LLMs to automate survey generation for better efficiency. However, the quality gap between LLM-generated surveys and those written by human remains significant, particularly in terms of outline quality and citation accuracy. To close these gaps, we introduce SurveyForge, which first generates the outline by analyzing the logical structure of human-written outlines and referring to the retrieved domain-related articles. Subsequently, leveraging high-quality papers retrieved from memory by our scholar navigation agent, SurveyForge can automatically generate and refine the content of the generated article. Moreover, to achieve a comprehensive evaluation, we construct SurveyBench, which includes 100 human-written survey papers for win-rate comparison and assesses AI-generated survey papers across three dimensions: reference, outline, and content quality. Experiments demonstrate that SurveyForge can outperform previous works such as AutoSurvey.

cs.CL↗

arXiv2025

$^{14}$N$/^{15}$N abundance ratio toward massive star-forming regions with different Galactic distances

The abundance ratio of $^{14}$N$/^{15}$N is, in principle, a powerful tool for tracing stellar nucleosynthesis. This work aims to measure and analyze ($^{14}$N/$^{15}$N)$\times$($^{13}$C/$^{12}$C) and $^{14}$N$/^{15}$N abundance ratios in massive star-forming regions across a range of galactocentric distances to provide constraints on galactic chemical evolution (GCE) models. We present H$^{13}$CN and HC$^{15}$N J=2-1 results toward 51massive star-forming regions obtained with the Institut de Radioastronomie Millimétrique (IRAM) 30 meter telescope.\ We used these results to derive ($^{14}$N/$^{15}$N)$\times$($^{13}$C/$^{12}$C) abundance ratios as well as $^{14}$N$/^{15}$N ratios using the double isotope method. We find an overall decreasing trend in the ($^{14}$N/$^{15}$N)$\times$($^{13}$C/$^{12}$C) abundance ratio and an increasing trend in the $^{14}$N$/^{15}$N ratio with increasing galactocentric distance ($D_{\rm GC}$), which provides a good constraint for the GCE model based on high signal to noise ratio measurements. While the predicted ($^{14}$N/$^{15}$N)$\times$($^{13}$C/$^{12}$C) ratios between 6 and 12 kpc determined using current GCE models are consistent with our observational results, the ratios from models for $D_{\rm GC}$ less than 6 kpc are significantly higher than the observational results, which indicates GCE models for $^{14}$N/$^{15}$N and/or $^{13}$C/$^{12}$C ratios need to be updated for at least this range.

astro-ph.GA↗

arXiv2025

Evaluating Intelligence via Trial and Error

Intelligence is a crucial trait for species to find solutions within a limited number of trial-and-error attempts. Building on this idea, we introduce Survival Game as a framework to evaluate intelligence based on the number of failed attempts in a trial-and-error process. Fewer failures indicate higher intelligence. When the expectation and variance of failure counts are both finite, it signals the ability to consistently find solutions to new challenges, which we define as the Autonomous Level of intelligence. Using Survival Game, we comprehensively evaluate existing AI systems. Our results show that while AI systems achieve the Autonomous Level in simple tasks, they are still far from it in more complex tasks, such as vision, search, recommendation, and language. While scaling current AI technologies might help, this would come at an astronomical cost. Projections suggest that achieving the Autonomous Level for general tasks would require $10^{26}$ parameters. To put this into perspective, loading such a massive model requires so many H100 GPUs that their total value is $10^{7}$ times that of Apple Inc.'s market value. Even with Moore's Law, supporting such a parameter scale would take $70$ years. This staggering cost highlights the complexity of human tasks and the inadequacies of current AI technologies. To further investigate this phenomenon, we conduct a theoretical analysis of Survival Game and its experimental results. Our findings suggest that human tasks possess a criticality property. As a result, Autonomous Level requires a deep understanding of the task's underlying mechanisms. Current AI systems, however, do not fully grasp these mechanisms and instead rely on superficial mimicry, making it difficult for them to reach an autonomous level. We believe Survival Game can not only guide the future development of AI but also offer profound insights into human intelligence.

cs.AI↗

arXiv2025

Empathy Level Alignment via Reinforcement Learning for Empathetic Response Generation

Empathetic response generation, aiming to understand the user's situation and feelings and respond empathically, is crucial in building human-like dialogue systems. Traditional approaches typically employ maximum likelihood estimation as the optimization objective during training, yet fail to align the empathy levels between generated and target responses. To this end, we propose an empathetic response generation framework using reinforcement learning (EmpRL). The framework develops an effective empathy reward function and generates empathetic responses by maximizing the expected reward through reinforcement learning. EmpRL utilizes the pre-trained T5 model as the generator and further fine-tunes it to initialize the policy. To align the empathy levels between generated and target responses within a given context, an empathy reward function containing three empathy communication mechanisms -- emotional reaction, interpretation, and exploration -- is constructed using pre-designed and pre-trained empathy identifiers. During reinforcement learning training, the proximal policy optimization algorithm is used to fine-tune the policy, enabling the generation of empathetic responses. Both automatic and human evaluations demonstrate that the proposed EmpRL framework significantly improves the quality of generated responses, enhances the similarity in empathy levels between generated and target responses, and produces empathetic responses covering both affective and cognitive aspects.

cs.CL↗

arXiv2025

Dynamically Local-Enhancement Planner for Large-Scale Autonomous Driving

Current autonomous vehicles operate primarily within limited regions, but there is increasing demand for broader applications. However, as models scale, their limited capacity becomes a significant challenge for adapting to novel scenarios. It is increasingly difficult to improve models for new situations using a single monolithic model. To address this issue, we introduce the concept of dynamically enhancing a basic driving planner with local driving data, without permanently modifying the planner itself. This approach, termed the Dynamically Local-Enhancement (DLE) Planner, aims to improve the scalability of autonomous driving systems without significantly expanding the planner's size. Our approach introduces a position-varying Markov Decision Process formulation coupled with a graph neural network that extracts region-specific driving features from local observation data. The learned features describe the local behavior of the surrounding objects, which is then leveraged to enhance a basic reinforcement learning-based policy. We evaluated our approach in multiple scenarios and compared it with a one-for-all driving model. The results show that our method outperforms the baseline policy in both safety (collision rate) and average reward, while maintaining a lighter scale. This approach has the potential to benefit large-scale autonomous vehicles without the need for largely expanding on-device driving models.

cs.RO↗