arXiv subjects

Zishuo Xu

Publications and source records attributed to Zishuo Xu.

4 recordsLinked to original sources

arXiv2025

Self-Evolving 3D Scene Generation from a Single Image

Generating high-quality, textured 3D scenes from a single image remains a fundamental challenge in vision and graphics. Recent image-to-3D generators recover reasonable geometry from single views, but their object-centric training limits generalization to complex, large-scale scenes with faithful structure and texture. We present EvoScene, a self-evolving, training-free framework that progressively reconstructs complete 3D scenes from single images. The key idea is combining the complementary strengths of existing models: geometric reasoning from 3D generation models and visual knowledge from video generation models. Through three iterative stages--Spatial Prior Initialization, Visual-guided 3D Scene Mesh Generation, and Spatial-guided Novel View Generation--EvoScene alternates between 2D and 3D domains, gradually improving both structure and appearance. Experiments on diverse scenes demonstrate that EvoScene achieves superior geometric stability, view-consistent textures, and unseen-region completion compared to strong baselines, producing ready-to-use 3D meshes for practical applications.

cs.CV↗

arXiv2025

From Ranking to Reasoning: Explainable Web API Recommendation via Semantic Reasoning

The rapid growth of Web APIs has made automated Web API recommendation essential for efficient mashup development. However, existing approaches suffer from two major limitations: 1) they rely on fixed top-N recommendation strategies that cannot adapt to mashup complexity, and 2) they provide little or no explanation for recommended APIs, limiting transparency and user trust. To address these challenges, we propose WAR-R1, an explainable Web API recommendation framework that integrates semantic reasoning with adaptive, variable-cardinality recommendation. Built on a lightweight large language model (LLM), WAR-R1 generates both a set of relevant APIs and a natural-language justification for each recommendation. To support adaptive recommendation size, we introduce special start and stop tokens that allow the model to learn when to begin and terminate API generation. WAR-R1 is trained in two stages: supervised fine-tuning on an annotated mashup-API corpus, followed by reinforcement learning using Group Relative Policy Optimization (GRPO) with low-rank adaptation to jointly optimize recommendation accuracy and reasoning quality. Experiments on the ProgrammableWeb dataset show that WAR-R1 outperforms state-of-the-art baselines by up to 10.89% in recommendation accuracy while consistently producing high-quality, semantically grounded explanations. Extensive ablation studies validate the effectiveness of reinforcement learning, special token design, and integrated reasoning.

cs.SE↗

arXiv2025

WARBERT: A Hierarchical BERT-based Model for Web API Recommendation

With the rise of Web 2.0 and microservices, the increasing availability of Web APIs has intensified the need for effective recommendation systems. Existing approaches are generally categorized into two methods: recommendation-type methods, which classify APIs using labels, and match-type methods, which retrieve APIs through matching with mashups. However, three significant challenges remain: 1) semantic ambiguities in comparing API and mashup descriptions, 2) a lack of progressive semantic refinement between mashup requirements and individual API descriptions, and 3) computational inefficiency of exhaustive mashup-API comparisons in large-scale repositories. To tackle these challenges, we propose WARBERT, a hierarchical model based on BERT for Web API recommendation. WARBERT utilizes dual-component feature fusion and attention mechanisms to create accurate semantic representations. It consists of WARBERT(R) for initial candidate filtering using recommendation methods, and WARBERT(M), which focuses on refined similarity matching. The final likelihood of an API-mashup pairing combines predictions from both components, with WARBERT(R) further enhanced by an auxiliary task of predicting mashup categories. Experiments conducted on the ProgrammableWeb dataset demonstrate WARBERT outperforms existing baselines, achieving notable improvements in both accuracy and efficiency.

cs.IR↗

arXiv2025

Constructing a 3D Scene from a Single Image

Acquiring detailed 3D scenes typically demands costly equipment, multi-view data, or labor-intensive modeling. Therefore, a lightweight alternative, generating complex 3D scenes from a single top-down image, plays an essential role in real-world applications. While recent 3D generative models have achieved remarkable results at the object level, their extension to full-scene generation often leads to inconsistent geometry, layout hallucinations, and low-quality meshes. In this work, we introduce SceneFuse-3D, a training-free framework designed to synthesize coherent 3D scenes from a single top-down view. Our method is grounded in two principles: region-based generation to improve image-to-3D alignment and resolution, and spatial-aware 3D inpainting to ensure global scene coherence and high-quality geometry generation. Specifically, we decompose the input image into overlapping regions and generate each using a pretrained 3D object generator, followed by a masked rectified flow inpainting process that fills in missing geometry while maintaining structural continuity. This modular design allows us to overcome resolution bottlenecks and preserve spatial structure without requiring 3D supervision or fine-tuning. Extensive experiments across diverse scenes show that SceneFuse-3D outperforms state-of-the-art baselines, including Trellis, Hunyuan3D-2, TripoSG, and LGM, in terms of geometry quality, spatial coherence, and texture fidelity. Our results demonstrate that high-quality coherent 3D scene-level asset generation is achievable from a single top-down image using a principled, training-free pipeline.

cs.CV↗