arXiv subjects

Wei Zeng

Publications and source records attributed to Wei Zeng.

At least 37 records · Page 2Linked to original sources

arXiv2026

InkIdeator: Supporting Chinese-Style Visual Design Ideation via AI-Infused Exploration of Chinese Paintings

Visual designers often seek inspiration from Chinese paintings when tasked with creating Chinese-style illustrations, posters, etc. Our formative study (N=10) reveals that during ideation, designers learn the cultural symbols, emotions, compositions, and styles in Chinese paintings but face challenges in searching, analyzing, and integrating these dimensions. This paper leverages multi-modal large models to annotate the value of each dimension in 16,315 Chinese paintings, built on which we propose InkIdeator, an ideation support system for Chinese-style visual designs. InkIdeator suggests cultural symbols associated with the task theme, provides dimensional keywords to help analyze Chinese paintings, and generates visual examples integrating user-selected keywords. Our within-subjects study (N=12) using a baseline system without extracted dimensional keywords, along with two extended use cases by Chinese painters, indicates InkIdeator's effectiveness in creative ideation support, helping users efficiently explore cultural dimensions in Chinese paintings and visualize their ideas. We discuss implications for supporting culture-related visual design ideation with generative AI.

cs.HC↗

arXiv2026

nvBench 2.0: Resolving Ambiguity in Text-to-Visualization through Stepwise Reasoning

Text-to-Visualization (Text2VIS) enables users to create visualizations from natural language queries, making data insights more accessible. However, Text2VIS faces challenges in interpreting ambiguous queries, as users often express their visualization needs in imprecise language. To address this challenge, we introduce nBench 2.0, a new benchmark designed to evaluate Text2VIS systems in scenarios involving ambiguous queries. nvBench 2.0 includes 7,878 natural language queries and 24,076 corresponding visualizations, derived from 780 tables across 153 domains. It is built using a controlled ambiguity-injection pipeline that generates ambiguous queries through a reverse-generation workflow. By starting with unambiguous seed visualizations and selectively injecting ambiguities, the pipeline yields multiple valid interpretations for each query, with each ambiguous query traceable to its corresponding visualization through step-wise reasoning paths. We evaluate various Large Language Models (LLMs) on their ability to perform ambiguous Text2VIS tasks using nBench 2.0. We also propose Step-Text2Vis, an LLM-based model trained on nvBench 2.0, which enhances performance in ambiguous scenarios through step-wise preference optimization. Our results show that Step-Text2Vis outperforms all baselines, setting a new state-of-the-art for ambiguous Text2VIS tasks. Our source code and data are available at https://nvbench2.github.io/

cs.CL↗

arXiv2025

GeoReasoner: Geo-localization with Reasoning in Street Views using a Large Vision-Language Model

This work tackles the problem of geo-localization with a new paradigm using a large vision-language model (LVLM) augmented with human inference knowledge. A primary challenge here is the scarcity of data for training the LVLM - existing street-view datasets often contain numerous low-quality images lacking visual clues, and lack any reasoning inference. To address the data-quality issue, we devise a CLIP-based network to quantify the degree of street-view images being locatable, leading to the creation of a new dataset comprising highly locatable street views. To enhance reasoning inference, we integrate external knowledge obtained from real geo-localization games, tapping into valuable human inference capabilities. The data are utilized to train GeoReasoner, which undergoes fine-tuning through dedicated reasoning and location-tuning stages. Qualitative and quantitative evaluations illustrate that GeoReasoner outperforms counterpart LVLMs by more than 25% at country-level and 38% at city-level geo-localization tasks, and surpasses StreetCLIP performance while requiring fewer training resources. The data and code are available at https://github.com/lingli1996/GeoReasoner.

cs.CV↗

arXiv2025

C2Views: Knowledge-based Colormap Design for Multiple-View Consistency

Multiple-view (MV) visualization provides a comprehensive and integrated perspective on complex data, establishing itself as an effective method for visual communication and exploratory data analysis. While existing studies have predominantly focused on designing explicit visual linkages and coordinated interactions to facilitate the exploration of MV visualizations, these approaches often demand extra graphical and interactive effort, overlooking the potential of color as an effective channel for encoding data and relationships. Addressing this oversight, we introduce C2Views, a new framework for colormap design that implicitly shows the relation across views. We begin by structuring the components and their relationships within MVs into a knowledge-based graph specification, wherein colormaps, data, and views are denoted as entities, and the interactions among them are illustrated as relations. Building on this representation, we formulate the design criteria as an optimization problem and employ a genetic algorithm enhanced by Pareto optimality, generating colormaps that balance single-view effectiveness and multiple-view consistency. Our approach is further complemented with an interactive interface for user-intended refinement. We demonstrate the feasibility of C2Views through various colormap design examples for MVs, underscoring its adaptability to diverse data relationships and view layouts. Comparative user studies indicate that our method outperforms the existing approach in facilitating color distinction and enhancing multiple-view consistency, thereby simplifying data exploration processes.

cs.HC↗

arXiv2025

When Eyes and Ears Disagree: Can MLLMs Discern Audio-Visual Confusion?

Can Multimodal Large Language Models (MLLMs) discern confused objects that are visually present but audio-absent? To study this, we introduce a new benchmark, AV-ConfuseBench, which simulates an ``Audio-Visual Confusion'' scene by modifying the corresponding sound of an object in the video, e.g., mute the sounding object and ask MLLMs Is there a/an muted-object sound''. Experimental results reveal that MLLMs, such as Qwen2.5-Omni and Gemini 2.5, struggle to discriminate non-existent audio due to visually dominated reasoning. Motivated by this observation, we introduce RL-CoMM, a Reinforcement Learning-based Collaborative Multi-MLLM that is built upon the Qwen2.5-Omni foundation. RL-CoMM includes two stages: 1) To alleviate visually dominated ambiguities, we introduce an external model, a Large Audio Language Model (LALM), as the reference model to generate audio-only reasoning. Then, we design a Step-wise Reasoning Reward function that enables MLLMs to self-improve audio-visual reasoning with the audio-only reference. 2) To ensure an accurate answer prediction, we introduce Answer-centered Confidence Optimization to reduce the uncertainty of potential heterogeneous reasoning differences. Extensive experiments on audio-visual question answering and audio-visual hallucination show that RL-CoMM improves the accuracy by 10~30\% over the baseline model with limited training data. Follow: https://github.com/rikeilong/AVConfusion.

cs.CV↗

arXiv2025

Peransformer: Improving Low-informed Expressive Performance Rendering with Score-aware Discriminator

Highly-informed Expressive Performance Rendering (EPR) systems transform music scores with rich musical annotations into human-like expressive performance MIDI files. While these systems have achieved promising results, the availability of detailed music scores is limited compared to MIDI files and are less flexible to work with using a digital audio workstation (DAW). Recent advancements in low-informed EPR systems offer a more accessible alternative by directly utilizing score-derived MIDI as input, but these systems often exhibit suboptimal performance. Meanwhile, existing works are evaluated with diverse automatic metrics and data formats, hindering direct objective comparisons between EPR systems. In this study, we introduce Peransformer, a transformer-based low-informed EPR system designed to bridge the gap between low-informed and highly-informed EPR systems. Our approach incorporates a score-aware discriminator that leverages the underlying score-derived MIDI files and is trained on a score-to-performance paired, note-to-note aligned MIDI dataset. Experimental results demonstrate that Peransformer achieves state-of-the-art performance among low-informed systems, as validated by subjective evaluations. Furthermore, we extend existing automatic evaluation metrics for EPR systems and introduce generalized EPR metrics (GEM), enabling more direct, accurate, and reliable comparisons across EPR systems.

cs.SD↗

arXiv2025

Disentangling Score Content and Performance Style for Joint Piano Rendering and Transcription

Expressive performance rendering (EPR) and automatic piano transcription (APT) are fundamental yet inverse tasks in music information retrieval: EPR generates expressive performances from symbolic scores, while APT recovers scores from performances. Despite their dual nature, prior work has addressed them independently. In this paper we propose a unified framework that jointly models EPR and APT by disentangling note-level score content and global performance style representations from both paired and unpaired data. Our framework is built on a transformer-based sequence-to-sequence architecture and is trained using only sequence-aligned data, without requiring fine-grained note-level alignment. To automate the rendering process while ensuring stylistic compatibility with the score, we introduce an independent diffusion-based performance style recommendation module that generates style embeddings directly from score content. This modular component supports both style transfer and flexible rendering across a range of expressive styles. Experimental results from both objective and subjective evaluations demonstrate that our framework achieves competitive performance on EPR and APT tasks, while enabling effective content-style disentanglement, reliable style transfer, and stylistically appropriate rendering. Demos are available at https://jointpianist.github.io/epr-apt/

cs.SD↗

arXiv2025

Multi-level Value Alignment in Agentic AI Systems: Survey and Perspectives

The ongoing evolution of AI paradigms has propelled AI research into the agentic AI stage. Consequently, the focus of research has shifted from single agents and simple applications towards multi-agent autonomous decision-making and task collaboration in complex environments. As Large Language Models (LLMs) advance, their applications become more diverse and complex, leading to increasing situational and systemic risks. This has brought significant attention to value alignment for agentic AI systems, which aims to ensure that an agent's goals, preferences, and behaviors align with human values and societal norms. Addressing socio-governance demands through a Multi-level Value framework, this study comprehensively reviews value alignment in LLM-based multi-agent systems as the representative archetype of agentic AI systems. Our survey systematically examines three interconnected dimensions: First, value principles are structured via a top-down hierarchy across macro, meso, and micro levels. Second, application scenarios are categorized along a general-to-specific continuum explicitly mirroring these value tiers. Third, value alignment methods and evaluation are mapped to this tiered framework through systematic examination of benchmarking datasets and relevant methodologies. Additionally, we delve into value coordination among multiple agents within agentic AI systems. Finally, we propose several potential research directions in this field.

cs.AI↗

arXiv2025

MCCoder: Streamlining Motion Control with LLM-Assisted Code Generation and Rigorous Verification

Large Language Models (LLMs) have demonstrated significant potential in code generation. However, in the factory automation sector, particularly motion control, manual programming, alongside inefficient and unsafe debugging practices, remains prevalent. This stems from the complex interplay of mechanical and electrical systems and stringent safety requirements. Moreover, most current AI-assisted motion control programming efforts focus on PLCs, with little attention given to high-level languages and function libraries. To address these challenges, we introduce MCCoder, an LLM-powered system tailored for generating motion control code, integrated with a soft-motion controller. MCCoder improves code generation through a structured workflow that combines multitask decomposition, hybrid retrieval-augmented generation (RAG), and iterative self-correction, utilizing a well-established motion library. Additionally, it integrates a 3D simulator for intuitive motion validation and logs of full motion trajectories for data verification, significantly enhancing accuracy and safety. In the absence of benchmark datasets and metrics tailored for evaluating motion control code generation, we propose MCEVAL, a dataset spanning motion tasks of varying complexity. Experiments show that MCCoder outperforms baseline models using Advanced RAG, achieving an overall performance gain of 33.09% and a 131.77% improvement on complex tasks in the MCEVAL dataset.

cs.AI↗

arXiv2025

QC-OT: Optimal Transport with Quasiconformal Mapping

The optimal transport (OT) map offers the most economical way to transfer one probability measure distribution to another. Classical OT theory does not involve a discussion of preserving topological connections and orientations in transmission results and processes. Existing numerical and geometric methods for computing OT seldom pays specific attention on this aspect. Especially, when dealing with the triangular mesh data, the known semi-discrete geometric OT (sd-OT) method employs critical operation of Delaunay triangulation (DT) to adapt topology to ensure the convexity of the energy function and the existence of the solution. This change in topology hampers the applicability of OT in modeling non-flip physical deformations in real-world tasks such as shape registration and editing problems in computer vision and medical imaging fields. This work introduces the topology structure-preserving optimal transport (QC-OT) map for the triangular mesh input. The computational strategy focuses on the two components: relaxing DT and convexity check in sd-OT and integrating quasiconformal (QC) correction. Here, quasiconformal mapping is employed to correct the regions unexpected distortions, and guarantee the topological preserving property of the transport. Furthermore, the spatial-temporal topology-preserving OT map is presented based t-OT to study the dynamics of the transportation. Multiple experiments have validated the efficiency and effectiveness of the proposed method and demonstrated its potential in the applications of mesh parameterization and image editing.

math.GN↗

arXiv2025

NN-Former: Rethinking Graph Structure in Neural Architecture Representation

The growing use of deep learning necessitates efficient network design and deployment, making neural predictors vital for estimating attributes such as accuracy and latency. Recently, Graph Neural Networks (GNNs) and transformers have shown promising performance in representing neural architectures. However, each of both methods has its disadvantages. GNNs lack the capabilities to represent complicated features, while transformers face poor generalization when the depth of architecture grows. To mitigate the above issues, we rethink neural architecture topology and show that sibling nodes are pivotal while overlooked in previous research. We thus propose a novel predictor leveraging the strengths of GNNs and transformers to learn the enhanced topology. We introduce a novel token mixer that considers siblings, and a new channel mixer named bidirectional graph isomorphism feed-forward network. Our approach consistently achieves promising performance in both accuracy and latency prediction, providing valuable insights for learning Directed Acyclic Graph (DAG) topology. The code is available at https://github.com/XuRuihan/NNFormer.

cs.LG↗

arXiv2025

AKRMap: Adaptive Kernel Regression for Trustworthy Visualization of Cross-Modal Embeddings

Cross-modal embeddings form the foundation for multi-modal models. However, visualization methods for interpreting cross-modal embeddings have been primarily confined to traditional dimensionality reduction (DR) techniques like PCA and t-SNE. These DR methods primarily focus on feature distributions within a single modality, whilst failing to incorporate metrics (e.g., CLIPScore) across multiple modalities. This paper introduces AKRMap, a new DR technique designed to visualize cross-modal embeddings metric with enhanced accuracy by learning kernel regression of the metric landscape in the projection space. Specifically, AKRMap constructs a supervised projection network guided by a post-projection kernel regression loss, and employs adaptive generalized kernels that can be jointly optimized with the projection. This approach enables AKRMap to efficiently generate visualizations that capture complex metric distributions, while also supporting interactive features such as zoom and overlay for deeper exploration. Quantitative experiments demonstrate that AKRMap outperforms existing DR methods in generating more accurate and trustworthy visualizations. We further showcase the effectiveness of AKRMap in visualizing and comparing cross-modal embeddings for text-to-image models. Code and demo are available at https://github.com/yilinye/AKRMap.

cs.CV↗

arXiv2025

Machine learning-based parameter optimization for Müntz spectral methods

Spectral methods employing non-standard polynomial bases, such as Müntz polynomials, have proven effective for accurately solving problems with solutions exhibiting low regularity, notably including sub-diffusion equations. However, due to the absence of theoretical guidance, the key parameters controlling the exponents of Müntz polynomials are usually determined empirically through extensive numerical experiments, leading to a time-consuming tuning process. To address this issue, we propose a novel machine learning-based optimization framework for the Müntz spectral method. As an illustrative example, we optimize the parameter selection for solving time-fractional partial differential equations (PDEs). Specifically, an artificial neural network (ANN) is employed to predict optimal parameter values based solely on the time-fractional order as input. The ANN is trained by minimizing solution errors on a one-dimensional time-fractional convection-diffusion equation featuring manufactured exact solutions that manifest singularities of varying intensity, covering a comprehensive range of sampled fractional orders. Numerical results for time-fractional PDEs in both one and two dimensions demonstrate that the ANN-based parameter prediction significantly improves the accuracy of the Müntz spectral method. Moreover, the trained ANN generalizes effectively from one-dimensional to two-dimensional cases, highlighting its robustness across spatial dimensions. Additionally, we verify that the ANN substantially outperforms traditional function approximators, such as spline interpolation, in both prediction accuracy and training efficiency. The proposed optimization framework can be extended beyond fractional PDEs, offering a versatile and powerful approach for spectral methods applied to various low-regularity problems.

math.NA↗

arXiv2025

Facilitating Video Story Interaction with Multi-Agent Collaborative System

Video story interaction enables viewers to engage with and explore narrative content for personalized experiences. However, existing methods are limited to user selection, specially designed narratives, and lack customization. To address this, we propose an interactive system based on user intent. Our system uses a Vision Language Model (VLM) to enable machines to understand video stories, combining Retrieval-Augmented Generation (RAG) and a Multi-Agent System (MAS) to create evolving characters and scene experiences. It includes three stages: 1) Video story processing, utilizing VLM and prior knowledge to simulate human understanding of stories across three modalities. 2) Multi-space chat, creating growth-oriented characters through MAS interactions based on user queries and story stages. 3) Scene customization, expanding and visualizing various story scenes mentioned in dialogue. Applied to the Harry Potter series, our study shows the system effectively portrays emergent character social behavior and growth, enhancing the interactive experience in the video story world.

cs.HC↗

arXiv2025

VizTA: Enhancing Comprehension of Distributional Visualization with Visual-Lexical Fused Conversational Interface

Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.

cs.HC↗

arXiv2025

Multi-modal Reference Learning for Fine-grained Text-to-Image Retrieval

Fine-grained text-to-image retrieval aims to retrieve a fine-grained target image with a given text query. Existing methods typically assume that each training image is accurately depicted by its textual descriptions. However, textual descriptions can be ambiguous and fail to depict discriminative visual details in images, leading to inaccurate representation learning. To alleviate the effects of text ambiguity, we propose a Multi-Modal Reference learning framework to learn robust representations. We first propose a multi-modal reference construction module to aggregate all visual and textual details of the same object into a comprehensive multi-modal reference. The multi-modal reference hence facilitates the subsequent representation learning and retrieval similarity computation. Specifically, a reference-guided representation learning module is proposed to use multi-modal references to learn more accurate visual and textual representations. Additionally, we introduce a reference-based refinement method that employs the object references to compute a reference-based similarity that refines the initial retrieval results. Extensive experiments are conducted on five fine-grained text-to-image retrieval datasets for different text-to-image retrieval tasks. The proposed method has achieved superior performance over state-of-the-art methods. For instance, on the text-to-person image retrieval dataset RSTPReid, our method achieves the Rank1 accuracy of 56.2\%, surpassing the recent CFine by 5.6\%.

cs.CV↗

arXiv2025

GMapLatent: Geometric Mapping in Latent Space

Cross-domain generative models based on encoder-decoder AI architectures have attracted much attention in generating realistic images, where domain alignment is crucial for generation accuracy. Domain alignment methods usually deal directly with the initial distribution; however, mismatched or mixed clusters can lead to mode collapse and mixture problems in the decoder, compromising model generalization capabilities. In this work, we innovate a cross-domain alignment and generation model that introduces a canonical latent space representation based on geometric mapping to align the cross-domain latent spaces in a rigorous and precise manner, thus avoiding mode collapse and mixture in the encoder-decoder generation architectures. We name this model GMapLatent. The core of the method is to seamlessly align latent spaces with strict cluster correspondence constraints using the canonical parameterizations of cluster-decorated latent spaces. We first (1) transform the latent space to a canonical parameter domain by composing barycenter translation, optimal transport merging and constrained harmonic mapping, and then (2) compute geometric registration with cluster constraints over the canonical parameter domains. This process realizes a bijective (one-to-one and onto) mapping between newly transformed latent spaces and generates a precise alignment of cluster pairs. Cross-domain generation is then achieved through the aligned latent spaces embedded in the encoder-decoder pipeline. Experiments on gray-scale and color images validate the efficiency, efficacy and applicability of GMapLatent, and demonstrate that the proposed model has superior performance over existing models.

cs.CV↗

arXiv2025

GenColor: Generative Color-Concept Association in Visual Design

Existing approaches for color-concept association typically rely on query-based image referencing, and color extraction from image references. However, these approaches are effective only for common concepts, and are vulnerable to unstable image referencing and varying image conditions. Our formative study with designers underscores the need for primary-accent color compositions and context-dependent colors (e.g., 'clear' vs. 'polluted' sky) in design. In response, we introduce a generative approach for mining semantically resonant colors leveraging images generated by text-to-image models. Our insight is that contemporary text-to-image models can resemble visual patterns from large-scale real-world data. The framework comprises three stages: concept instancing produces generative samples using diffusion models, text-guided image segmentation identifies concept-relevant regions within the image, and color association extracts primarily accompanied by accent colors. Quantitative comparisons with expert designs validate our approach's effectiveness, and we demonstrate the applicability through cases in various design scenarios and a gallery.

cs.HC↗