Shuang Zhao

Cheng Sun, Guangyan Cai, Zhengqin Li et al. · nvidia

Reconstructing the shape and spatially varying surface appearances of a physical-world object as well as its surrounding illumination based on 2D images (e.g., photographs) of the object has been a long-standing problem in computer vision and graphics. In this paper, we introduce an accurate and highly efficient object reconstruction pipeline combining neural based object reconstruction and physics-based inverse rendering (PBIR). Our pipeline firstly leverages a neural SDF based shape reconstruction to produce high-quality but potentially imperfect object shape. Then, we introduce a neural material and lighting distillation stage to achieve high-quality predictions for material and illumination. In the last stage, initialized by the neural predictions, we perform PBIR to refine the initial results and obtain the final high-quality reconstruction of object shape, material, and illumination. Experimental results demonstrate our pipeline significantly outperforms existing methods quality-wise and performance-wise.

Zihan Liu, Jiaqi Wang, Yun Luo et al.

In recent years, there has been an explosion of research on the application of deep learning to the prediction of various peptide properties, due to the significant development and market potential of peptides. Molecular dynamics has enabled the efficient collection of large peptide datasets, providing reliable training data for deep learning. However, the lack of systematic analysis of the peptide encoding, which is essential for AI-assisted peptide-related tasks, makes it an urgent problem to be solved for the improvement of prediction accuracy. To address this issue, we first collect a high-quality, colossal simulation dataset of peptide self-assembly containing over 62,000 samples generated by coarse-grained molecular dynamics (CGMD). Then, we systematically investigate the effect of peptide encoding of amino acids into sequences and molecular graphs using state-of-the-art sequential (i.e., RNN, LSTM, and Transformer) and structural deep learning models (i.e., GCN, GAT, and GraphSAGE), on the accuracy of peptide self-assembly prediction, an essential physiochemical process prior to any peptide-related applications. Extensive benchmarking studies have proven Transformer to be the most powerful sequence-encoding-based deep learning model, pushing the limit of peptide self-assembly prediction to decapeptides. In summary, this work provides a comprehensive benchmark analysis of peptide encoding with advanced deep learning models, serving as a guide for a wide range of peptide-related predictions such as isoelectric points, hydration free energy, etc.

Bowen Xue, Shuang Zhao, Henrik Wann Jensen et al.

Neural reflectance models are capable of reproducing the spatially-varying appearance of many real-world materials at different scales. Unfortunately, existing techniques such as NeuMIP have difficulties handling materials with strong shadowing effects or detailed specular highlights. In this paper, we introduce a neural appearance model that offers a new level of accuracy. Central to our model is an inception-based core network structure that captures material appearances at multiple scales using parallel-operating kernels and ensures multi-stage features through specialized convolution layers. Furthermore, we encode the inputs into frequency space, introduce a gradient-based loss, and employ it adaptive to the progress of the learning phase. We demonstrate the effectiveness of our method using a variety of synthetic and real examples.

Kai Yan, Fujun Luan, MiloŠ HaŠAn et al.

A 3D digital scene contains many components: lights, materials and geometries, interacting to reach the desired appearance. Staging such a scene is time-consuming and requires both artistic and technical skills. In this work, we propose PSDR-Room, a system allowing to optimize lighting as well as the pose and materials of individual objects to match a target image of a room scene, with minimal user input. To this end, we leverage a recent path-space differentiable rendering approach that provides unbiased gradients of the rendering with respect to geometry, lighting, and procedural materials, allowing us to optimize all of these components using gradient descent to visually match the input photo appearance. We use recent single-image scene understanding methods to initialize the optimization and search for appropriate 3D models and materials. We evaluate our method on real photographs of indoor scenes and demonstrate the editability of the resulting scene components.

Jun Lv, Yunhai Feng, Cheng Zhang et al.

Model-based reinforcement learning (MBRL) is recognized with the potential to be significantly more sample-efficient than model-free RL. How an accurate model can be developed automatically and efficiently from raw sensory inputs (such as images), especially for complex environments and tasks, is a challenging problem that hinders the broad application of MBRL in the real world. In this work, we propose a sensing-aware model-based reinforcement learning system called SAM-RL. Leveraging the differentiable physics-based simulation and rendering, SAM-RL automatically updates the model by comparing rendered images with real raw images and produces the policy efficiently. With the sensing-aware learning pipeline, SAM-RL allows a robot to select an informative viewpoint to monitor the task process. We apply our framework to real world experiments for accomplishing three manipulation tasks: robotic assembly, tool manipulation, and deformable object manipulation. We demonstrate the effectiveness of SAM-RL via extensive experiments. Videos are available on our project webpage at https://sites.google.com/view/rss-sam-rl.

Guangyan Cai, Fujun Luan, Miloš Hašan et al.

Glossy objects present a significant challenge for 3D reconstruction from multi-view input images under natural lighting. In this paper, we introduce PBIR-NIE, an inverse rendering framework designed to holistically capture the geometry, material attributes, and surrounding illumination of such objects. We propose a novel parallax-aware non-distant environment map as a lightweight and efficient lighting representation, accurately modeling the near-field background of the scene, which is commonly encountered in real-world capture setups. This feature allows our framework to accommodate complex parallax effects beyond the capabilities of standard infinite-distance environment maps. Our method optimizes an underlying signed distance field (SDF) through physics-based differentiable rendering, seamlessly connecting surface gradients between a triangle mesh and the SDF via neural implicit evolution (NIE). To address the intricacies of highly glossy BRDFs in differentiable rendering, we integrate the antithetic sampling algorithm to mitigate variance in the Monte Carlo gradient estimator. Consequently, our framework exhibits robust capabilities in handling glossy object reconstruction, showcasing superior quality in geometry, relighting, and material estimation.

Siyuan Yan, Xieji Li, Dan Mo et al.

Medical foundation models have shown promise in controlled benchmarks, yet widespread deployment remains hindered by reliance on task-specific fine-tuning. Here, we introduce DermFM-Zero, a dermatology vision-language foundation model trained via masked latent modelling and contrastive learning on over 4 million multimodal data points. We evaluated DermFM-Zero across 20 benchmarks spanning zero-shot diagnosis and multimodal retrieval, achieving state-of-the-art performance without task-specific adaptation. We further evaluated its zero-shot capabilities in three multinational reader studies involving over 1,100 clinicians. In primary care settings, AI assistance enabled general practitioners to nearly double their differential diagnostic accuracy across 98 skin conditions. In specialist settings, the model significantly outperformed board-certified dermatologists in multimodal skin cancer assessment. In collaborative workflows, AI assistance enabled non-experts to surpass unassisted experts while improving management appropriateness. Finally, we show that DermFM-Zero's latent representations are interpretable: sparse autoencoders unsupervisedly disentangle clinically meaningful concepts that outperform predefined-vocabulary approaches and enable targeted suppression of artifact-induced biases, enhancing robustness without retraining. These findings demonstrate that a foundation model can provide effective, safe, and transparent zero-shot clinical decision support.

Yan-Jie Zhou, Wei Liu, Yuan Gao et al.

Skin diseases are among the most prevalent health issues, and accurate computer-aided diagnosis methods are of importance for both dermatologists and patients. However, most of the existing methods overlook the essential domain knowledge required for skin disease diagnosis. A novel multi-task model, namely DermImitFormer, is proposed to fill this gap by imitating dermatologists' diagnostic procedures and strategies. Through multi-task learning, the model simultaneously predicts body parts and lesion attributes in addition to the disease itself, enhancing diagnosis accuracy and improving diagnosis interpretability. The designed lesion selection module mimics dermatologists' zoom-in action, effectively highlighting the local lesion features from noisy backgrounds. Additionally, the presented cross-interaction module explicitly models the complicated diagnostic reasoning between body parts, lesion attributes, and diseases. To provide a more robust evaluation of the proposed method, a large-scale clinical image dataset of skin diseases with significantly more cases than existing datasets has been established. Extensive experiments on three different datasets consistently demonstrate the state-of-the-art recognition performance of the proposed approach.

Peiyu Xu, Xin Sun, Krishna Mullia et al.

Ray-tracing-based 3D Gaussian splatting (3DGS) methods overcome the limitations of rasterization -- rigid pinhole camera assumptions, inaccurate shadows, and lack of native reflection or refraction -- but remain slower due to the cost of sorting all intersecting Gaussians along every ray. Moreover, existing ray-tracing methods still rely on rasterization-style approximations such as shadow mapping for relightable scenes, undermining the generality that ray tracing promises. We present a differentiable, sorting-free stochastic formulation for ray-traced 3DGS -- the first framework that uses stochastic ray tracing to both reconstruct and render standard and relightable 3DGS scenes. At its core is an unbiased Monte Carlo estimator for pixel-color gradients that evaluates only a small sampled subset of Gaussians per ray, bypassing the need for sorting. For standard 3DGS, our method matches the reconstruction quality and speed of rasterization-based 3DGS while substantially outperforming sorting-based ray tracing. For relightable 3DGS, the same stochastic estimator drives per-Gaussian shading with fully ray-traced shadow rays, delivering notably higher reconstruction fidelity than prior work.

Jingwang Ling, Ruihan Yu, Feng Xu et al.

Physics-based inverse rendering enables joint optimization of shape, material, and lighting based on captured 2D images. To ensure accurate reconstruction, using a light model that closely resembles the captured environment is essential. Although the widely adopted distant environmental lighting model is adequate in many cases, we demonstrate that its inability to capture spatially varying illumination can lead to inaccurate reconstructions in many real-world inverse rendering scenarios. To address this limitation, we incorporate NeRF as a non-distant environment emitter into the inverse rendering pipeline. Additionally, we introduce an emitter importance sampling technique for NeRF to reduce the rendering variance. Through comparisons on both real and synthetic datasets, our results demonstrate that our NeRF-based emitter offers a more precise representation of scene lighting, thereby improving the accuracy of inverse rendering.

Bowen Xue, Giuseppe Claudio Guarnera, Shuang Zhao et al.

We introduce Reflectance Diffusion, a new neural text-to-texture model capable of generating high-fidelity SVBRDF maps from textual descriptions. Our method leverages a tandem neural approach, consisting of two modules, to accurately model the distribution of spatially varying reflectance as described by text prompts. Initially, we employ a pre-trained stable diffusion 2 model to generate a latent representation that informs the overall shape of the material and serves as our backbone model. Then, our ReflectanceUNet enables fine-tuning control over the material's physical appearance and generates SVBRDF maps. ReflectanceUNet module is trained on an extensive dataset comprising approximately 200,000 synthetic spatially varying materials. Our generative SVBRDF diffusion model allows for the synthesis of multiple SVBRDF estimates from a single textual input, offering users the possibility to choose the output that best aligns with their requirements. We illustrate our method's versatility by generating SVBRDF maps from a range of textual descriptions, both specific and broad. Our ReflectanceUNet model can integrate optional physical parameters, such as roughness and specularity, enhancing customization. When the backbone module is fixed, the ReflectanceUNet module refines the material, allowing direct edits to its physical attributes. Comparative evaluations demonstrate that ReflectanceFusion achieves better accuracy than existing text-to-material models, such as Text2Mat, while also providing the benefits of editable and relightable SVBRDF maps.

Hua Cai, Shuang Zhao, Liang Zhang et al.

Reasoning-focused large language models (LLMs) are rapidly evolving across various domains, yet their capabilities in handling complex legal problems remains underexplored. In this paper, we introduce Unilaw-R1, a large language model tailored for legal reasoning. With a lightweight 7-billion parameter scale, Unilaw-R1 significantly reduces deployment cost while effectively tackling three core challenges in the legal domain: insufficient legal knowledge, unreliable reasoning logic, and weak business generalization. To address these issues, we first construct Unilaw-R1-Data, a high-quality dataset containing 17K distilled and screened chain-of-thought (CoT) samples. Based on this, we adopt a two-stage training strategy combining Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), which significantly boosts the performance on complex legal reasoning tasks and supports interpretable decision-making in legal AI applications. To assess legal reasoning ability, we also introduce Unilaw-R1-Eval, a dedicated benchmark designed to evaluate models across single- and multi-choice legal tasks. Unilaw-R1 demonstrates strong results on authoritative benchmarks, outperforming all models of similar scale and achieving performance on par with the much larger DeepSeek-R1-Distill-Qwen-32B (54.9%). Following domain-specific training, it also showed significant gains on LawBench and LexEval, exceeding Qwen-2.5-7B-Instruct (46.6%) by an average margin of 6.6%.

Bowen Xue, Giuseppe Claudio Guarnera, Shuang Zhao et al.

Current video diffusion models generate visually compelling content but often violate basic laws of physics, producing subtle artifacts like rubber-sheet deformations and inconsistent object motion. We introduce a frequency-domain physics prior that improves motion plausibility without modifying model architectures. Our method decomposes common rigid motions (translation, rotation, scaling) into lightweight spectral losses, requiring only 2.7% of frequency coefficients while preserving 97%+ of spectral energy. Applied to Open-Sora, MVDIT, and Hunyuan, our approach improves both motion accuracy and action recognition by ~11% on average on OpenVID-1M (relative), while maintaining visual quality. User studies show 74--83% preference for our physics-enhanced videos. It also reduces warping error by 22--37% (depending on the backbone) and improves temporal consistency scores. These results indicate that simple, global spectral cues are an effective drop-in regularizer for physically plausible motion in video diffusion.

Chen Zhu, Liang Du, Hong Chen et al.

Click-Through Rate (CTR) prediction is a pivotal task in product and content recommendation, where learning effective feature embeddings is of great significance. However, traditional methods typically learn fixed feature representations without dynamically refining feature representations according to the context information, leading to suboptimal performance. Some recent approaches attempt to address this issue by learning bit-wise weights or augmented embeddings for feature representations, but suffer from uninformative or redundant features in the context. To tackle this problem, inspired by the Global Workspace Theory in conscious processing, which posits that only a specific subset of the product features are pertinent while the rest can be noisy and even detrimental to human-click behaviors, we propose a CTR model that enables Dynamic Embedding Learning with Truncated Conscious Attention for CTR prediction, termed DELTA. DELTA contains two key components: (I) conscious truncation module (CTM), which utilizes curriculum learning to apply adaptive truncation on attention weights to select the most critical feature in the context; (II) explicit embedding optimization (EEO), which applies an auxiliary task during training that directly and independently propagates the gradient from the loss layer to the embedding layer, thereby optimizing the embedding explicitly via linear feature crossing. Extensive experiments on five challenging CTR datasets demonstrate that DELTA achieves new state-of-art performance among current CTR methods.

Shuang Zhao, Shuhui Chen, Ziling Wei

With the popularity of smartphones, mobile applications (apps) have penetrated the daily life of people. Although apps provide rich functionalities, they also access a large amount of personal information simultaneously. As a result, privacy concerns are raised. To understand what personal information the apps collect, many solutions are presented to detect privacy leaks in apps. Recently, the traffic monitoring-based privacy leak detection method has shown promising performance and strong scalability. However, it still has some shortcomings. Firstly, it suffers from detecting the leakage of personal information with obfuscation. Secondly, it cannot discover the privacy leaks of undefined type. Aiming at solving the above problems, a new personal information detection method based on traffic monitoring is proposed in this paper. In this paper, statistical features of personal information are designed to depict the occurrence patterns of personal information in the traffic, including local patterns and global patterns. Then a detector is trained based on machine learning algorithms to discover potential personal information with similar patterns. Since the statistical features are independent of the value and type of personal information, the trained detector is capable of identifying various types of privacy leaks and obfuscated privacy leaks. As far as we know, this is the first work that detects personal information based on statistical features. Finally, the experimental results show that the proposed method could achieve better performance than the state-of-the-art.

Fujun Luan, Shuang Zhao, Kavita Bala et al.

Reconstructing the shape and appearance of real-world objects using measured 2D images has been a long-standing problem in computer vision. In this paper, we introduce a new analysis-by-synthesis technique capable of producing high-quality reconstructions through robust coarse-to-fine optimization and physics-based differentiable rendering. Unlike most previous methods that handle geometry and reflectance largely separately, our method unifies the optimization of both by leveraging image gradients with respect to both object reflectance and geometry. To obtain physically accurate gradient estimates, we develop a new GPU-based Monte Carlo differentiable renderer leveraging recent advances in differentiable rendering theory to offer unbiased gradients while enjoying better performance than existing tools like PyTorch3D and redner. To further improve robustness, we utilize several shape and material priors as well as a coarse-to-fine optimization strategy to reconstruct geometry. We demonstrate that our technique can produce reconstructions with higher quality than previous methods such as COLMAP and Kinect Fusion.

Yu Guo, Cameron Smith, Miloš Hašan et al.

We address the problem of reconstructing spatially-varying BRDFs from a small set of image measurements. This is a fundamentally under-constrained problem, and previous work has relied on using various regularization priors or on capturing many images to produce plausible results. In this work, we present MaterialGAN, a deep generative convolutional network based on StyleGAN2, trained to synthesize realistic SVBRDF parameter maps. We show that MaterialGAN can be used as a powerful material prior in an inverse rendering framework: we optimize in its latent representation to generate material maps that match the appearance of the captured images when rendered. We demonstrate this framework on the task of reconstructing SVBRDFs from images captured under flash illumination using a hand-held mobile phone. Our method succeeds in producing plausible material maps that accurately reproduce the target images, and outperforms previous state-of-the-art material capture methods in evaluations on both synthetic and real data. Furthermore, our GAN-based latent space allows for high-level semantic material editing operations such as generating material variations and material morphing.

Yu Guo, Milos Hasan, Lingqi Yan et al.

Procedural material models have been gaining traction in many applications thanks to their flexibility, compactness, and easy editability. We explore the inverse rendering problem of procedural material parameter estimation from photographs, presenting a unified view of the problem in a Bayesian framework. In addition to computing point estimates of the parameters by optimization, our framework uses a Markov Chain Monte Carlo approach to sample the space of plausible material parameters, providing a collection of plausible matches that a user can choose from, and efficiently handling both discrete and continuous model parameters. To demonstrate the effectiveness of our framework, we fit procedural models of a range of materials---wall plaster, leather, wood, anisotropic brushed metals and layered metallic paints---to both synthetic and real target images.

Zixun Sun, Shuang Zhao, Chengwei Zhu et al.

Online personalized news product needs a suitable cover for the article. The news cover demands to be with high image quality, and draw readers' attention at same time, which is extraordinary challenging due to the subjectivity of the task. In this paper, we assess the news cover from image clarity and object salience perspective. We propose an end-to-end multi-task learning network for image clarity assessment and semantic segmentation simultaneously, the results of which can be guided for news cover assessment. The proposed network is based on a modified DeepLabv3+ model. The network backbone is used for multiple scale spatial features exaction, followed by two branches for image clarity assessment and semantic segmentation, respectively. The experiment results show that the proposed model is able to capture important content in images and performs better than single-task learning baselines on our proposed game content based CIA dataset.

Chengqian Che, Fujun Luan, Shuang Zhao et al.

We introduce inverse transport networks as a learning architecture for inverse rendering problems where, given input image measurements, we seek to infer physical scene parameters such as shape, material, and illumination. During training, these networks are evaluated not only in terms of how close they can predict groundtruth parameters, but also in terms of whether the parameters they produce can be used, together with physically-accurate graphics renderers, to reproduce the input image measurements. To en- able training of inverse transport networks using stochastic gradient descent, we additionally create a general-purpose, physically-accurate differentiable renderer, which can be used to estimate derivatives of images with respect to arbitrary physical scene parameters. Our experiments demonstrate that inverse transport networks can be trained efficiently using differentiable rendering, and that they generalize to scenes with completely unseen geometry and illumination better than networks trained without appearance- matching regularization.