SHShi-Min Hu

Shi-Min Hu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Recognitions

Top Cited Scholar (2024)

Shi-Min Hu has a total of 354 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    End-to-End Differentiable Smoothed Particle Hydrodynamics for Fluid--Rigid Optimization with Gradient Stabilization

    by Geng-Chen Cao, Bo Ren, Yanrui Xu, Zhe-Han Mo, Shi-Min Hu

    We introduce an end-to-end differentiable optimization framework for Smoothed Particle Hydrodynamics (SPH) with fluid–rigid body coupling. The central difficulty is not only to obtain local derivatives of SPH operators, but also to propagate useful sensitivities through long rollouts whose particle neighborhoods, contacts, and pressure responses change rapidly over time. These effects make reverse-mode gradients noisy and often cause adjoint variables to explode in long-horizon control and inverse-design tasks. We formulate an SPH-specific reverse-mode pipeline for IISPH-based fluid–rigid simulation and implement the required vector–Jacobian products within checkpointed simulation segments. To make the resulting gradients usable for optimization, we introduce a self-adaptive adjoint correction sequence that acts as a descent-preserving preconditioner: it bounds pathological adjoint amplification while keeping the update direction aligned with the original reverse-mode sensitivity. Our experiments cover a sequence of increasingly demanding tasks, from low-dimensional impulse and shape optimization to scalar-amplitude water actuation, multi-object water-jet balancing, and neural fluid-mediated pendulum control. These results demonstrate that stabilized reverse-mode differentiation can support challenging SPH-based optimization problems involving indirect momentum transfer, long-horizon event timing, and high-dimensional control.

  • Open Access

    SkinTokens : A Learned Compact Representation for Unified Autoregressive Rigging

    by Jiapeng Zhang, Chengfeng Pu, Meng-Hao Guo, Yan-Pei Cao, Shi-Min Hu

    The rapid proliferation of generative 3D models has created a critical bottleneck in animation pipelines: rigging. Existing automated methods are fundamentally limited by their approach to skinning, treating it as an ill-posed, high-dimensional regression task that is inefficient to optimize and is typically decoupled from skeleton generation. We posit this is a representation problem and introduce SkinTokens : a learned, compact, and discrete representation for skinning weights. By leveraging an FSQ-CVAE to capture the intrinsic sparsity of skinning, we reframe the task from continuous regression to a more tractable token sequence prediction problem. This representation enables TokenRig , a unified autoregressive framework that models the entire rig as a single sequence of skeletal parameters and SkinTokens, learning the complicated dependencies between skeletons and skin deformations. The unified model is then amenable to a reinforcement learning stage, where tailored geometric and semantic rewards improve generalization to complex, out-of-distribution assets. Quantitatively, the SkinTokens representation leads to a 98%–133% relative improvement in skinning-weight L 1 error over state-of-the-art methods, while the full TokenRig framework, refined with RL, improves the skeletal Chamfer distance by 17%–22% . Our work presents a unified, generative approach to rigging that yields higher fidelity and robustness, offering a scalable solution to a long-standing challenge in 3D content creation.

  • Open Access

    PackMonitor: Towards Zero Package Hallucinations via Decoding-Time Monitoring

    by Xiting Liu, Yuetong Liu, Yitong Zhang, Jia Li, Shi-Min Hu

    As Large Language Models (LLMs) are increasingly integrated into software development workflows, their trustworthiness has become a critical concern. However, in dependency recommendation scenarios, the reliability of LLMs is undermined by widespread package hallucinations, where models often recommend hallucinated packages. Recent studies have proposed a range of approaches to mitigate this issue. Nevertheless, existing approaches typically merely reduce hallucination rates rather than eliminate them, leaving persistent software security risks. In this work, we argue that package hallucinations are theoretically preventable based on the key insight that package validity is decidable through finite and enumerable authoritative package lists. Building on this, we propose PackMonitor, the first approach capable of fundamentally eliminating package hallucinations by continuously monitoring the model's decoding process and intervening when necessary. To implement this in practice, PackMonitor addresses three key challenges: (1) determining when to trigger intervention via a Context-Aware Parser that continuously monitors model outputs and selectively activates intervening only during installation command generation; (2) resolving how to intervene by employing a Package-Name Intervenor that strictly limits the decoding space to an authoritative package list; and (3) ensuring monitoring efficiency through a DFA-Caching Mechanism that enables scalability to millions of packages with negligible overhead. Extensive experiments on five widely used LLMs demonstrate that PackMonitor is a training-free, plug-and-play solution that consistently reduces package hallucination rates to zero while maintaining low-latency inference and preserving original model capabilities.

  • Beyond Inpainting: Unleash 3D Understanding for Stable Camera-Controlled Video Re-Rendering

    by Dong-Yu Chen, Yixin Guo, Shuojin Yang, Tai-Jiang Mu, Shi-Min Hu

  • Open Access

    GPR: Towards a Generative Pre-trained One-Model Paradigm for Large-Scale Advertising Recommendation

    by Jie Jiang, Jun Zhang, Yi Li, Yue Liu, Changping Wang, Yuan Wang, Yuling Xiong, Xun Liu, Haiyang Wu, Qian Li, et al.

    Online advertising recommendation is the economic engine of the digital ecosystem, yet traditional multi-stage cascading systems suffer from objective misalignment and error propagation. Conversely, emerging generative models often struggle with the strict latency and revenue alignment requirements of industrial applications. To address these issues, we propose GPR (Generative Pre-trained Recommender). Instead of the complex retrieval and pre-ranking cascade, GPR employs a unified generative framework to directly generate candidate sets, which are then refined by a lightweight fine-ranking stage for precise business alignment. Unlike prior works, GPR introduces three key technical innovations to ensure industrial feasibility: (1) Unified Representation: We utilize RQ-KMeans+ to map heterogeneous user behaviors and ad content into a shared semantic ID space; (2) Decoupled Architecture: We design a Heterogeneous Hierarchical Decoder (HHD) to decouple user intent modeling from ad generation, enabling efficient long-horizon inference; (3) Value-Aligned Generation: We propose a Value-Guided Trie-Constrained Beam Search and a hierarchical training strategy (MTP + HEPO) to strictly enforce business constraints and maximize commercial value. GPR has been fully deployed in Tencent Weixin Channels, serving over 500 million daily active users. Extensive online A/B testing demonstrates that GPR delivers substantial gains in GMV and CTCVR over highly optimized cascading baselines, validating the generative paradigm as a superior solution for large-scale advertising systems.

  • Open Access

    Pixal3D: Pixel-Aligned 3D Generation from Images

    by Dong-Yang Li, Wang Zhao, Yuxin Chen, Wenbo Hu, Meng-Hao Guo, Fang-Lue Zhang, Ying Shan, Shi-Min Hu


  • Open Access

    TopoCap: Learning Topology-Agnostic Motion Priors for Monocular Video-to-Animation

    by Cheng-Feng Pu, Jia-Peng Zhang, Meng-Hao Guo, Yan-Pei Cao, Shi-Min Hu

  • Spatial Multiple Importance Sampling for Real-Time Irradiance Probes

    by Tuo Chen, Zi-Heng Zhou, Lingqi Yan, Shi-Min Hu

    Real-time global illumination rendered with low variance remains a persistent challenge. Many engines employ irradiance probes as a relatively cheap technique, but constrained computational budgets often lead to flickering artifacts in rendered images. In this paper, we propose spatial multiple importance sampling, which reuses ray-surface intersection data among real-time irradiance probes to significantly reduce flicker caused by variance, enabling efficient computation in complex scenes under limited ray tracing budgets. Moreover, our approach incorporates a probe selection mechanism to enhance reuse efficiency and a visibility estimation method to mitigate bias. Experimental results demonstrate that our method significantly reduces variance at a fixed ray tracing cost, delivering high-quality, stable outputs in real-time scenarios.

  • SeparateGen: Semantic Component-Based 3D Character Generation From Single Images

    by Dong-Yang Li, Yi-Long Liu, Zi-Xian Liu, Yan-Pei Cao, Meng-Hao Guo, Shi-Min Hu

    Creating detailed 3D characters from a single image remains challenging due to the difficulty in separating semantic components during generation. Existing methods often produce entangled meshes with poor topology, hindering downstream applications like rigging and animation. We introduce SeparateGen, a novel framework that generates high-quality 3D characters by explicitly reconstructing them as distinct semantic components (e.g., body, clothing, hair, shoes) from a single, arbitrary-pose image. SeparateGen first leverages a multi-view diffusion model to generate consistent multi-view images in a canonical A-pose. Then, a novel component-aware reconstruction model, SC-LRM, conditioned on these multi-view images, adaptively decomposes and reconstructs each component with high fidelity. To train and evaluate SeparateGen, we contribute SC-Anime, the first large-scale dataset of 7,580 anime-style 3D characters with detailed component-level annotations. Extensive experiments demonstrate that SeparateGen significantly outperforms state-of-the-art methods in both reconstruction quality and multi-view consistency. Furthermore, our component-based approach effectively resolves mesh entanglement issues, enabling seamless rigging and asset reuse. SeparateGen thus represents a step towards generating high-quality, application-ready 3D characters from a single image. The SC-Anime dataset and our code will be publicly released.