Search

Capturing the temporal evolution of Gaussian properties such as position, rotation, and scale is a challenging task due to the vast number of time-varying parameters and the limited photometric data available, which generally results in convergence issues, making it difficult to find an optimal solution. While feeding all inputs into an end-to-end neural network can effectively model complex temporal dynamics, this approach lacks explicit supervision and struggles to generate high-quality transformation fields. On the other hand, using time-conditioned polynomial functions to model Gaussian trajectories and orientations provides a more explicit and interpretable solution, but requires significant handcrafted effort and lacks generalizability across diverse scenes. To overcome these limitations, this paper introduces a novel approach based on a learnable infinite Taylor Formula to model the temporal evolution of Gaussians. This method offers both the flexibility of an implicit network-based approach and the interpretability of explicit polynomial functions, allowing for more robust and generalizable modeling of Gaussian dynamics across various dynamic scenes. Extensive experiments on dynamic novel view rendering tasks are conducted on public datasets, demonstrating that the proposed method achieves state-of-the-art performance in this domain. More information is available on our project page(https://ellisonking.github.io/TaylorGaussian).

In the past few decades, autonomous driving algorithms have made significant progress in perception, planning, and control. However, evaluating individual components does not fully reflect the performance of entire systems, highlighting the need for more holistic assessment methods. This motivates the development of HUGSIM, a closed-loop, photo-realistic, and real-time simulator for evaluating autonomous driving algorithms. We achieve this by lifting captured 2D RGB images into the 3D space via 3D Gaussian Splatting, improving the rendering quality for closed-loop scenarios, and building the closed-loop environment. In terms of rendering, We tackle challenges of novel view synthesis in closed-loop scenarios, including viewpoint extrapolation and 360-degree vehicle rendering. Beyond novel view synthesis, HUGSIM further enables the full closed simulation loop, dynamically updating the ego and actor states and observations based on control commands. Moreover, HUGSIM offers a comprehensive benchmark across more than 70 sequences from KITTI-360, Waymo, nuScenes, and PandaSet, along with over 400 varying scenarios, providing a fair and realistic evaluation platform for existing autonomous driving algorithms. HUGSIM not only serves as an intuitive evaluation benchmark but also unlocks the potential for fine-tuning autonomous driving algorithms in a photorealistic closed-loop setting., Comment: Our project page is at https://xdimlab.github.io/HUGSIM

Recent numerous video generation models, also known as world models, have demonstrated the ability to generate plausible real-world videos. However, many studies have shown that these models often produce motion results lacking logical or physical coherence. In this paper, we revisit video generation models and find that single-stage approaches struggle to produce high-quality results while maintaining coherent motion reasoning. To address this issue, we propose \textbf{Motion Dreamer}, a two-stage video generation framework. In Stage I, the model generates an intermediate motion representation-such as a segmentation map or depth map-based on the input image and motion conditions, focusing solely on the motion itself. In Stage II, the model uses this intermediate motion representation as a condition to generate a high-detail video. By decoupling motion reasoning from high-fidelity video synthesis, our approach allows for more accurate and physically plausible motion generation. We validate the effectiveness of our approach on the Physion dataset and in autonomous driving scenarios. For example, given a single push, our model can synthesize the sequential toppling of a set of dominoes. Similarly, by varying the movements of ego-cars, our model can produce different effects on other vehicles. Our work opens new avenues in creating models that can reason about physical interactions in a more coherent and realistic manner.

Standard clothing asset generation involves creating forward-facing flat-lay garment images displayed on a clear background by extracting clothing information from diverse real-world contexts, which presents significant challenges due to highly standardized sampling distributions and precise structural requirements in the generated images. Existing models have limited spatial perception and often exhibit structural hallucinations in this high-specification generative task. To address this issue, we propose a novel Retrieval-Augmented Generation (RAG) framework, termed RAGDiffusion, to enhance structure determinacy and mitigate hallucinations by assimilating external knowledge from LLM and databases. RAGDiffusion consists of two core processes: (1) Retrieval-based structure aggregation, which employs contrastive learning and a Structure Locally Linear Embedding (SLLE) to derive global structure and spatial landmarks, providing both soft and hard guidance to counteract structural ambiguities; and (2) Omni-level faithful garment generation, which introduces a three-level alignment that ensures fidelity in structural, pattern, and decoding components within the diffusing. Extensive experiments on challenging real-world datasets demonstrate that RAGDiffusion synthesizes structurally and detail-faithful clothing assets with significant performance improvements, representing a pioneering effort in high-specification faithful generation with RAG to confront intrinsic hallucinations and enhance fidelity., Comment: Project website: https://colorful-liyu.github.io/RAGDiffusion-page/

This technical report summarizes the second-place solution for the Predictive World Model Challenge held at the CVPR-2024 Workshop on Foundation Models for Autonomous Systems. We introduce D$^2$-World, a novel World model that effectively forecasts future point clouds through Decoupled Dynamic flow. Specifically, the past semantic occupancies are obtained via existing occupancy networks (e.g., BEVDet). Following this, the occupancy results serve as the input for a single-stage world model, generating future occupancy in a non-autoregressive manner. To further simplify the task, dynamic voxel decoupling is performed in the world model. The model generates future dynamic voxels by warping the existing observations through voxel flow, while remaining static voxels can be easily obtained through pose transformation. As a result, our approach achieves state-of-the-art performance on the OpenScene Predictive World Model benchmark, securing second place, and trains more than 300% faster than the baseline model. Code is available at https://github.com/zhanghm1995/D2-World., Comment: The 2nd Place and Innovation Award Solution of Predictive World Model at the CVPR 2024 Autonomous Grand Challenge

Urban scene reconstruction is crucial for real-world autonomous driving simulators. Although existing methods have achieved photorealistic reconstruction, they mostly focus on pinhole cameras and neglect fisheye cameras. In fact, how to effectively simulate fisheye cameras in driving scene remains an unsolved problem. In this work, we propose UniGaussian, a novel approach that learns a unified 3D Gaussian representation from multiple camera models for urban scene reconstruction in autonomous driving. Our contributions are two-fold. First, we propose a new differentiable rendering method that distorts 3D Gaussians using a series of affine transformations tailored to fisheye camera models. This addresses the compatibility issue of 3D Gaussian splatting with fisheye cameras, which is hindered by light ray distortion caused by lenses or mirrors. Besides, our method maintains real-time rendering while ensuring differentiability. Second, built on the differentiable rendering method, we design a new framework that learns a unified Gaussian representation from multiple camera models. By applying affine transformations to adapt different camera models and regularizing the shared Gaussians with supervision from different modalities, our framework learns a unified 3D Gaussian representation with input data from multiple sources and achieves holistic driving scene understanding. As a result, our approach models multiple sensors (pinhole and fisheye cameras) and modalities (depth, semantic, normal and LiDAR point clouds). Our experiments show that our method achieves superior rendering quality and fast rendering speed for driving scene simulation., Comment: Technical report

This paper introduces VisionPAD, a novel self-supervised pre-training paradigm designed for vision-centric algorithms in autonomous driving. In contrast to previous approaches that employ neural rendering with explicit depth supervision, VisionPAD utilizes more efficient 3D Gaussian Splatting to reconstruct multi-view representations using only images as supervision. Specifically, we introduce a self-supervised method for voxel velocity estimation. By warping voxels to adjacent frames and supervising the rendered outputs, the model effectively learns motion cues in the sequential data. Furthermore, we adopt a multi-frame photometric consistency approach to enhance geometric perception. It projects adjacent frames to the current frame based on rendered depths and relative poses, boosting the 3D geometric representation through pure image supervision. Extensive experiments on autonomous driving datasets demonstrate that VisionPAD significantly improves performance in 3D object detection, occupancy prediction and map segmentation, surpassing state-of-the-art pre-training strategies by a considerable margin.

Confidence calibration in LLMs, i.e., aligning their self-assessed confidence with the actual accuracy of their responses, enabling them to self-evaluate the correctness of their outputs. However, current calibration methods for LLMs typically estimate two scalars to represent overall response confidence and correctness, which is inadequate for long-form generation where the response includes multiple atomic facts and may be partially confident and correct. These methods also overlook the relevance of each fact to the query. To address these challenges, we propose a Fact-Level Calibration framework that operates at a finer granularity, calibrating confidence to relevance-weighted correctness at the fact level. Furthermore, comprehensive analysis under the framework inspired the development of Confidence-Guided Fact-level Self-Correction ($\textbf{ConFix}$), which uses high-confidence facts within a response as additional knowledge to improve low-confidence ones. Extensive experiments across four datasets and six models demonstrate that ConFix effectively mitigates hallucinations without requiring external knowledge sources such as retrieval systems., Comment: Code is available at https://github.com/yuanyige/fact-calibration

The cameras equipped on mobile terminals employ different sensors in different photograph modes, and the transferability of raw domain denoising models between these sensors is significant but remains sufficient exploration. Industrial solutions either develop distinct training strategies and models for different sensors or ignore the differences between sensors and simply extend existing models to new sensors, which leads to tedious training or unsatisfactory performance. In this paper, we introduce a new benchmark, the Multi-Sensor SIDD (MSSIDD) dataset, which is the first raw-domain dataset designed to evaluate the sensor transferability of denoising models. The MSSIDD dataset consists of 60,000 raw images of six distinct sensors, derived through the degeneration of sRGB images via different camera sensor parameters. Furthermore, we propose a sensor consistency training framework that enables denoising models to learn the sensor-invariant features, thereby facilitating the generalization of the consistent model to unseen sensors. We evaluate previous arts on the newly proposed MSSIDD dataset, and the experimental results validate the effectiveness of our proposed method. Our dataset is available at https://www.kaggle.com/datasets/sjtuwh/mssidd., Comment: 15 pages,7 figures

Latent representation alignment has become a foundational technique for constructing multimodal large language models (MLLM) by mapping embeddings from different modalities into a shared space, often aligned with the embedding space of large language models (LLMs) to enable effective cross-modal understanding. While preliminary protein-focused MLLMs have emerged, they have predominantly relied on heuristic approaches, lacking a fundamental understanding of optimal alignment practices across representations. In this study, we explore the alignment of multimodal representations between LLMs and Geometric Deep Models (GDMs) in the protein domain. We comprehensively evaluate three state-of-the-art LLMs (Gemma2-2B, LLaMa3.1-8B, and LLaMa3.1-70B) with four protein-specialized GDMs (GearNet, GVP, ScanNet, GAT). Our work examines alignment factors from both model and protein perspectives, identifying challenges in current alignment methodologies and proposing strategies to improve the alignment process. Our key findings reveal that GDMs incorporating both graph and 3D structural information align better with LLMs, larger LLMs demonstrate improved alignment capabilities, and protein rarity significantly impacts alignment performance. We also find that increasing GDM embedding dimensions, using two-layer projection heads, and fine-tuning LLMs on protein-specific data substantially enhance alignment quality. These strategies offer potential enhancements to the performance of protein-related multimodal models. Our code and data are available at https://github.com/Tizzzzy/LLM-GDM-alignment., Comment: 24 pages, 9 figures

A very robust and efficient hybrid analytic-numerical Fock build, aMECP+aCOSx, has been developed for accelerating HF/DFT calculations. The essential idea is to extract those portions of the Fock matrix that can readily be evaluated analytically, so as to minimize numerical noises arising from the semi-numerical and numerical integrations. As a result, the combination of aMECP with a medium grid and aCOSx with a coarse grid is already sufficient to achieve an accuracy of less than 1{\mu}Eh/atom in total energies. The acceleration of aMECP+aCOSx over the analytic Fock build is already seen in calculations of small molecular systems and is more enhanced in calculations of large molecules with extended basis sets., Comment: 34 pages, 6 figures

Inspired by the would-be ${\eta}_{t}$ meson and promising prospects at the LHC, the ${\eta}_{t}$ ${\to}$ $W^{+}W^{-}$ decay is proposed to identify the ${\eta}_{t}$ meson. It is found that branching ratio for the ${\eta}_{t}$ ${\to}$ $W^{+}W^{-}$ decay is about $4.2{\times}10^{-5}$, and more than 800 events can be observed with the accumulated data of $2.7{\times}10^{7}$ ${\eta}_{t}$ meson at the HL-LHC in the future.

We investigate the dynamical property of the naive mean dimension for continuous actions of any countable group on compact metrizable spaces. It is shown that naive mean dimension serves as an upper bound of sofic mean dimension for actions of nonamenable groups. For algebraic actions we obtain more satisfactory results by looking at the naive version of mean rank for modules over integral group rings. We also consider the naive metric mean dimension and investigate its relations with sofic metric mean dimension., Comment: 25 pages. Comments are welcome!

Exploring new knowledge is a fundamental human ability that can be mirrored in the development of deep neural networks, especially in the field of object detection. Open world object detection (OWOD) is an emerging area of research that adapts this principle to explore new knowledge. It focuses on recognizing and learning from objects absent from initial training sets, thereby incrementally expanding its knowledge base when new class labels are introduced. This survey paper offers a thorough review of the OWOD domain, covering essential aspects, including problem definitions, benchmark datasets, source codes, evaluation metrics, and a comparative study of existing methods. Additionally, we investigate related areas like open set recognition (OSR) and incremental learning (IL), underlining their relevance to OWOD. Finally, the paper concludes by addressing the limitations and challenges faced by current OWOD algorithms and proposes directions for future research. To our knowledge, this is the first comprehensive survey of the emerging OWOD field with over one hundred references, marking a significant step forward for object detection technology. A comprehensive source code and benchmarks are archived and concluded at https://github.com/ArminLee/OWOD Review.

Human-object interaction (HOI) and human-scene interaction (HSI) are crucial for human-centric scene understanding applications in Embodied Artificial Intelligence (EAI), robotics, and augmented reality (AR). A common limitation faced in these research areas is the data scarcity problem: insufficient labeled human-scene object pairs on the input images, and limited interaction complexity and granularity between them. Recent HOI and HSI methods have addressed this issue by generating dynamic interactions with rigid objects. But more complex dynamic interactions such as a human rider pedaling an articulated bicycle have been unexplored. To address this limitation, and to enable research on complex dynamic human-articulated object interactions, in this paper we propose a method to generate simulated 3D dynamic cyclist assets and interactions. We designed a methodology for creating a new part-based multi-view articulated synthetic 3D bicycle dataset that we call 3DArticBikes that can be used to train NeRF and 3DGS-based 3D reconstruction methods. We then propose a 3DGS-based parametric bicycle composition model to assemble 8-DoF pose-controllable 3D bicycles. Finally, using dynamic information from cyclist videos, we build a complete synthetic dynamic 3D cyclist (rider pedaling a bicycle) by re-posing a selectable synthetic 3D person while automatically placing the rider onto one of our new articulated 3D bicycles using a proposed 3D Keypoint optimization-based Inverse Kinematics pose refinement. We present both, qualitative and quantitative results where we compare our generated cyclists against those from a recent stable diffusion-based method.

Test-Time Adaptation (TTA) has emerged as a promising paradigm for enhancing the generalizability of models. However, existing mainstream TTA methods, predominantly operating at batch level, often exhibit suboptimal performance in complex real-world scenarios, particularly when confronting outliers or mixed distributions. This phenomenon stems from a pronounced over-reliance on statistical patterns over the distinct characteristics of individual instances, resulting in a divergence between the distribution captured by the model and data characteristics. To address this challenge, we propose Meet-In-The-Middle based Test-Time Adaptation ($\textbf{MITA}$), which introduces energy-based optimization to encourage mutual adaptation of the model and data from opposing directions, thereby meeting in the middle. MITA pioneers a significant departure from traditional approaches that focus solely on aligning the model to the data, facilitating a more effective bridging of the gap between model's distribution and data characteristics. Comprehensive experiments with MITA across three distinct scenarios (Outlier, Mixture, and Pure) demonstrate its superior performance over SOTA methods, highlighting its potential to significantly enhance generalizability in practical applications.

Detecting covariate drift is a common task of significant practical value in supervised learning. Once covariate drift occurs, the models may no longer be applicable, hence numerous studies have been devoted to the advancement of detection methods. However, current research methods are not particularly effective in handling subtle covariate drift when dealing with small proportions of drift samples. In this paper, inspired by the $k$-nearest neighbor (KNN) approach, a novel method called Half $k$-farthest neighbor (Half-KFN) is proposed in response to specific scenarios. Compared to traditional ones, Half-KFN exhibits higher power due to the inherent capability of the farthest neighbors which could better characterize the nature of drift. Furthermore, with larger sample sizes, the employment of the bootstrap for hypothesis testing is recommended. It is leveraged to calculate $p$-values dramatically faster than permutation tests, with speed undergoing an exponential growth as sample size increases. Numerical experiments on simulated and real data are conducted to evaluate our proposed method, and the results demonstrate that it consistently displays superior sensitivity and rapidity in covariate drift detection across various cases.

Learning skills that interact with objects is of major importance for robotic manipulation. These skills can indeed serve as an efficient prior for solving various manipulation tasks. We propose a novel Skill Learning approach that discovers composable behaviors by solving a large and diverse number of autonomously generated tasks. Our method learns skills allowing the robot to consistently and robustly interact with objects in its environment. The discovered behaviors are embedded in primitives which can be composed with Hierarchical Reinforcement Learning to solve unseen manipulation tasks. In particular, we leverage Asymmetric Self-Play to discover behaviors and Multiplicative Compositional Policies to embed them. We compare our method to Skill Learning baselines and find that our skills are more interactive. Furthermore, the learned skills can be used to solve a set of unseen manipulation tasks, in simulation as well as on a real robotic platform., Comment: Accepted at the 2024 IEEE-RAS International Conference on Humanoid Robots

We present OmniBooth, an image generation framework that enables spatial control with instance-level multi-modal customization. For all instances, the multimodal instruction can be described through text prompts or image references. Given a set of user-defined masks and associated text or image guidance, our objective is to generate an image, where multiple objects are positioned at specified coordinates and their attributes are precisely aligned with the corresponding guidance. This approach significantly expands the scope of text-to-image generation, and elevates it to a more versatile and practical dimension in controllability. In this paper, our core contribution lies in the proposed latent control signals, a high-dimensional spatial feature that provides a unified representation to integrate the spatial, textual, and image conditions seamlessly. The text condition extends ControlNet to provide instance-level open-vocabulary generation. The image condition further enables fine-grained control with personalized identity. In practice, our method empowers users with more flexibility in controllable generation, as users can choose multi-modal conditions from text or images as needed. Furthermore, thorough experiments demonstrate our enhanced performance in image synthesis fidelity and alignment across different tasks and datasets. Project page: https://len-li.github.io/omnibooth-web/

The advancement of autonomous driving is increasingly reliant on high-quality annotated datasets, especially in the task of 3D occupancy prediction, where the occupancy labels require dense 3D annotation with significant human effort. In this paper, we propose SyntheOcc, which denotes a diffusion model that Synthesize photorealistic and geometric-controlled images by conditioning Occupancy labels in driving scenarios. This yields an unlimited amount of diverse, annotated, and controllable datasets for applications like training perception models and simulation. SyntheOcc addresses the critical challenge of how to efficiently encode 3D geometric information as conditional input to a 2D diffusion model. Our approach innovatively incorporates 3D semantic multi-plane images (MPIs) to provide comprehensive and spatially aligned 3D scene descriptions for conditioning. As a result, SyntheOcc can generate photorealistic multi-view images and videos that faithfully align with the given geometric labels (semantics in 3D voxel space). Extensive qualitative and quantitative evaluations of SyntheOcc on the nuScenes dataset prove its effectiveness in generating controllable occupancy datasets that serve as an effective data augmentation to perception models.

Leveraging the visual priors of pre-trained text-to-image diffusion models offers a promising solution to enhance zero-shot generalization in dense prediction tasks. However, existing methods often uncritically use the original diffusion formulation, which may not be optimal due to the fundamental differences between dense prediction and image generation. In this paper, we provide a systemic analysis of the diffusion formulation for the dense prediction, focusing on both quality and efficiency. And we find that the original parameterization type for image generation, which learns to predict noise, is harmful for dense prediction; the multi-step noising/denoising diffusion process is also unnecessary and challenging to optimize. Based on these insights, we introduce Lotus, a diffusion-based visual foundation model with a simple yet effective adaptation protocol for dense prediction. Specifically, Lotus is trained to directly predict annotations instead of noise, thereby avoiding harmful variance. We also reformulate the diffusion process into a single-step procedure, simplifying optimization and significantly boosting inference speed. Additionally, we introduce a novel tuning strategy called detail preserver, which achieves more accurate and fine-grained predictions. Without scaling up the training data or model capacity, Lotus achieves SoTA performance in zero-shot depth and normal estimation across various datasets. It also enhances efficiency, being significantly faster than most existing diffusion-based methods. Lotus' superior quality and efficiency also enable a wide range of practical applications, such as joint estimation, single/multi-view 3D reconstruction, etc. Project page: https://lotus3d.github.io/., Comment: The first two authors contributed equally. Project page: https://lotus3d.github.io/

Most of the known tight lower bounds for dynamic problems are based on the Online Boolean Matrix-Vector Multiplication (OMv) Hypothesis, which is not as well studied and understood as some more popular hypotheses in fine-grained complexity. It would be desirable to base hardness of dynamic problems on a more believable hypothesis. We propose analogues of the OMv Hypothesis for variants of matrix multiplication that are known to be harder than Boolean product in the offline setting, namely: equality, dominance, min-witness, min-max, and bounded monotone min-plus products. These hypotheses are a priori weaker assumptions than the standard (Boolean) OMv Hypothesis. Somewhat surprisingly, we show that they are actually equivalent to it. This establishes the first such fine-grained equivalence class for dynamic problems.

The allyl radical (C3H5) is a well-characterized hydrocarbon radical, renowned for its pivotal role as an intermediate species in high-energy environments. Its core excited states can elucidate intricate details pertaining to its electronic and structural properties. The core excited states of allyl were studied experimentally using X-ray absorption spectroscopy (XAS), and the primary characteristic peaks were assigned using the MCSCF approach, but not entirely. In this work, the recently developed CVS-icMRCISD scheme was used to simulate the excitation and ionization processes of C's K-shell electrons within allyl radicals, cations, and anions, respectively. Our results indicate that the XAS spectrum obtained not merely captured the distinctive peaks associated with allyl radicals, but also encompassed the characteristic peaks pertaining to allyl cations. Meanwhile, unlike manually adjusting the state weights of different electronic states to align with experimental spectral data, we adopt the CVS-icMRCISD scheme, which uses state averaging and produces unbiased results, making it suitable for studying multiple states simultaneously and easy to converge. More importantly, when accounting for the dynamic electron correlation, our results align seamlessly with the experimental XAS. This congruence underscores the potential of our CVS-icMRCISD as a robust tool for theoretical investigations pertaining to the excitation of inner shell electrons in small molecules.

Recently, an increasing number of news organizations have integrated artificial intelligence (AI) into their workflows, leading to a further influx of AI technologists and data workers into the news industry. This has initiated cross-functional collaborations between these professionals and journalists. While prior research has explored the impact of AI-related roles entering the news industry, there is a lack of studies on how cross-functional collaboration unfolds between AI professionals and journalists. Through interviews with 17 journalists, 6 AI technologists, and 3 AI workers with cross-functional experience from leading news organizations, we investigate the current practices, challenges, and opportunities for cross-functional collaboration around AI in today's news industry. We first study how journalists and AI professionals perceive existing cross-collaboration strategies. We further explore the challenges of cross-functional collaboration and provide recommendations for enhancing future cross-functional collaboration around AI in the news industry., Comment: 16 pages

Previous research pays attention to how users strategically understand and consciously interact with algorithms but mainly focuses on an individual level, making it difficult to explore how users within communities could develop a collective understanding of algorithms and organize collective algorithmic actions. Through a two-year ethnography of online fan activities, this study investigates 43 core fans who always organize large-scale fans' collective actions and their corresponding general fan groups. This study aims to reveal how these core fans mobilize millions of general fans through collective algorithmic actions. These core fans reported the rhetorical strategies used to persuade general fans, the steps taken to build a collective understanding of algorithms, and the collaborative processes that adapt collective actions across platforms and cultures. Our findings highlight the key factors that enable computer-supported collective algorithmic actions and extend collective action research into large-scale domain targeting algorithms., Comment: 15 pages

As an approximation to SDSCI [static-dynamic-static (SDS) configuration interaction (CI), a minimal MRCI; Theor. Chem. Acc. 133, 1481 (2014)], SDSPT2 [Mol. Phys. 115, 2696 (2017)] is a CI-like multireference (MR) second-order perturbation theory (PT2) that treats single and multiple roots on an equal footing. This feature permits the use of configuration selection over a large complete active space (CAS) $P$ to end up with a much reduced reference space $\tilde{P}$, which is connected only with a portion ($\tilde{Q}_1$) of the full first-order interacting space $Q$ connected to $P$. The effective interacting $\tilde{Q}$ space can further be truncated by an integral-based cutoff threshold. With marginal loss of accuracy, the selection-truncation procedure, along with an efficient evaluation and storage of internal contraction coefficients, renders SDSPT2s (SDSPT2 with selection) applicable to systems that cannot be handled by the parent CAS-based SDSPT2, as demonstrated by several challenging showcases., Comment: 52 pages, 7 figures

Accurate segmentation of rectal lymph nodes is crucial for the staging and treatment planning of rectal cancer. However, the complexity of the surrounding anatomical structures and the scarcity of annotated data pose significant challenges. This study introduces a novel lymph node synthesis technique aimed at generating diverse and realistic synthetic rectal lymph node samples to mitigate the reliance on manual annotation. Unlike direct diffusion methods, which often produce masks that are discontinuous and of suboptimal quality, our approach leverages an implicit SDF-based method for mask generation, ensuring the production of continuous, stable, and morphologically diverse masks. Experimental results demonstrate that our synthetic data significantly improves segmentation performance. Our work highlights the potential of diffusion model for accurately synthesizing structurally complex lesions, such as lymph nodes in rectal cancer, alleviating the challenge of limited annotated data in this field and aiding in advancements in rectal cancer diagnosis and treatment., Comment: 8 pages

3D Gaussian Splatting (3DGS) has attracted great attention in novel view synthesis because of its superior rendering efficiency and high fidelity. However, the trained Gaussians suffer from severe zooming degradation due to non-adjustable representation derived from single-scale training. Though some methods attempt to tackle this problem via post-processing techniques such as selective rendering or filtering techniques towards primitives, the scale-specific information is not involved in Gaussians. In this paper, we propose a unified optimization method to make Gaussians adaptive for arbitrary scales by self-adjusting the primitive properties (e.g., color, shape and size) and distribution (e.g., position). Inspired by the mipmap technique, we design pseudo ground-truth for the target scale and propose a scale-consistency guidance loss to inject scale information into 3D Gaussians. Our method is a plug-in module, applicable for any 3DGS models to solve the zoom-in and zoom-out aliasing. Extensive experiments demonstrate the effectiveness of our method. Notably, our method outperforms 3DGS in PSNR by an average of 9.25 dB for zoom-in and 10.40 dB for zoom-out on the NeRF Synthetic dataset., Comment: 9 pages

Stripe-like space target detection (SSTD) is crucial for space situational awareness. Traditional unsupervised methods often fail in low signal-to-noise ratio and variable stripe-like space targets scenarios, leading to weak generalization. Although fully supervised learning methods improve model generalization, they require extensive pixel-level labels for training. In the SSTD task, manually creating these labels is often inaccurate and labor-intensive. Semi-supervised learning (SSL) methods reduce the need for these labels and enhance model generalizability, but their performance is limited by pseudo-label quality. To address this, we introduce an innovative Collaborative Static-Dynamic Teacher (CSDT) SSL framework, which includes static and dynamic teacher models as well as a student model. This framework employs a customized adaptive pseudo-labeling (APL) strategy, transitioning from initial static teaching to adaptive collaborative teaching, guiding the student model's training. The exponential moving average (EMA) mechanism further enhances this process by feeding new stripe-like knowledge back to the dynamic teacher model through the student model, creating a positive feedback loop that continuously enhances the quality of pseudo-labels. Moreover, we present MSSA-Net, a novel SSTD network featuring a multi-scale dual-path convolution (MDPC) block and a feature map weighted attention (FMWA) block, designed to extract diverse stripe-like features within the CSDT SSL training framework. Extensive experiments verify the state-of-the-art performance of our framework on the AstroStripeSet and various ground-based and space-based real-world datasets.

Infrared small target sequences exhibit strong similarities between frames and contain rich contextual information, which motivates us to achieve sequential infrared small target segmentation (IRSTS) with minimal data. Inspired by the success of Segment Anything Model (SAM) across various downstream tasks, we propose a one-shot and training-free method that perfectly adapts SAM's zero-shot generalization capability to sequential IRSTS. Specifically, we first obtain a confidence map through local feature matching (LFM). The highest point in the confidence map is used as the prompt to replace the manual prompt. Then, to address the over-segmentation issue caused by the domain gap, we design the point prompt-centric focusing (PPCF) module. Subsequently, to prevent miss and false detections, we introduce the triple-level ensemble (TLE) module to produce the final mask. Experiments demonstrate that our method requires only one shot to achieve comparable performance to state-of-the-art IRSTS methods and significantly outperforms other one-shot segmentation methods. Moreover, ablation studies confirm the robustness of our method in the type of annotations and the selection of reference images.

Accurate behavior prediction for vehicles is essential but challenging for autonomous driving. Most existing studies show satisfying performance under regular scenarios, but most neglected safety-critical scenarios. In this study, a spatio-temporal dual-encoder network named STDA for safety-critical scenarios was developed. Considering the exceptional capabilities of human drivers in terms of situational awareness and comprehending risks, driver attention was incorporated into STDA to facilitate swift identification of the critical regions, which is expected to improve both performance and interpretability. STDA contains four parts: the driver attention prediction module, which predicts driver attention; the fusion module designed to fuse the features between driver attention and raw images; the temporary encoder module used to enhance the capability to interpret dynamic scenes; and the behavior prediction module to predict the behavior. The experiment data are used to train and validate the model. The results show that STDA improves the G-mean from 0.659 to 0.719 when incorporating driver attention and adopting a temporal encoder module. In addition, extensive experimentation has been conducted to validate that the proposed module exhibits robust generalization capabilities and can be seamlessly integrated into other mainstream models.

Driver action recognition, aiming to accurately identify drivers' behaviours, is crucial for enhancing driver-vehicle interactions and ensuring driving safety. Unlike general action recognition, drivers' environments are often challenging, being gloomy and dark, and with the development of sensors, various cameras such as IR and depth cameras have emerged for analyzing drivers' behaviors. Therefore, in this paper, we propose a novel multimodal fusion transformer, named MultiFuser, which identifies cross-modal interrelations and interactions among multimodal car cabin videos and adaptively integrates different modalities for improved representations. Specifically, MultiFuser comprises layers of Bi-decomposed Modules to model spatiotemporal features, with a modality synthesizer for multimodal features integration. Each Bi-decomposed Module includes a Modal Expertise ViT block for extracting modality-specific features and a Patch-wise Adaptive Fusion block for efficient cross-modal fusion. Extensive experiments are conducted on Drive&Act dataset and the results demonstrate the efficacy of our proposed approach.

For the short pulse initial data with a first order outgoing constraint condition and optimal orders of smallness, we establish the global existence of smooth solutions to 2D quasilinear wave equations with higher order null conditions. Such kinds of wave equations include 2D relativistic membrane equations, 2D membrane equations, and some 2D quasilinear equations which come from the nonlinear Maxwell equations in electromagnetic theory or from the corresponding Lagrangian functionals as perturbations of the Lagrangian densities of linear wave operators. The main ingredients of the analysis here include looking for a new good unknown, finding some key identities based on the higher order null conditions and the resulting null frames, as well as overcoming the difficulties due to the slow decay of solutions to the 2-D wave equation, so that the solutions can be estimated precisely.

Domain reweighting is an emerging research area aimed at adjusting the relative weights of different data sources to improve the effectiveness and efficiency of language model pre-training. This paper demonstrates that the optimal composition of training data from different domains is scale-dependent, challenging the existing practice of determining optimal mixtures through small-scale experiments and directly applying them at larger scales. We derive an analytical model for the dependence of optimal weights on data scale and introduce *AutoScale*, a novel, practical approach for optimizing data compositions at potentially large training data scales. *AutoScale* first uses a principled optimization framework to find optimal compositions at smaller, feasible scales, then predicts optimal compositions at larger scales using our derived model. Our evaluation on GPT-2 Large and BERT pre-training demonstrates *AutoScale*'s effectiveness in improving training convergence and downstream performance. Particularly, for GPT-2 Large on RedPajama, *AutoScale* decreases validation perplexity 28% faster than baselines, with up to 38% speed-up over unweighted training, achieving the best performance across downstream tasks. This work provides insights into the varying benefits of data sources across training scales for language models, contributing to the burgeoning research on scale-dependent data curation. Code is open-sourced., Comment: Preprint. Under review

Abstention, the refusal of large language models (LLMs) to provide an answer, is increasingly recognized for its potential to mitigate hallucinations and enhance safety in LLM systems. In this survey, we introduce a framework to examine abstention from three perspectives: the query, the model, and human values. We organize the literature on abstention methods, benchmarks, and evaluation metrics using this framework, and discuss merits and limitations of prior work. We further identify and motivate areas for future work, centered around whether abstention can be achieved as a meta-capability that transcends specific tasks or domains, while still providing opportunities to optimize abstention abilities based on context., Comment: preprint

Stripe-like space target detection (SSTD) plays a key role in enhancing space situational awareness and assessing spacecraft behaviour. This domain faces three challenges: the lack of publicly available datasets, interference from stray light and stars, and the variability of stripe-like targets, which makes manual labeling both inaccurate and labor-intensive. In response, we introduces `AstroStripeSet', a pioneering dataset designed for SSTD, aiming to bridge the gap in academic resources and advance research in SSTD. Furthermore, we propose a novel teacher-student label evolution framework with single-point weak supervision, providing a new solution to the challenges of manual labeling. This framework starts with generating initial pseudo-labels using the zero-shot capabilities of the Segment Anything Model (SAM) in a single-point setting. After that, the fine-tuned StripeSAM serves as the teacher and the newly developed StripeNet as the student, consistently improving segmentation performance through label evolution, which iteratively refines these labels. We also introduce `GeoDice', a new loss function customized for the linear characteristics of stripe-like targets. Extensive experiments show that our method matches fully supervised approaches, exhibits strong zero-shot generalization for diverse space-based and ground-based real-world images, and sets a new state-of-the-art (SOTA) benchmark. Our AstroStripeSet dataset and code will be made publicly available.

Trajectory prediction is significant for intelligent vehicles to achieve high-level autonomous driving, and a lot of relevant research achievements have been made recently. Despite the rapid development, most existing studies solely focused on normal safe scenarios while largely neglecting safety-critical scenarios, particularly those involving imminent collisions. This oversight may result in autonomous vehicles lacking the essential predictive ability in such situations, posing a significant threat to safety. To tackle these, this paper proposes a risk-aware trajectory prediction framework tailored to safety-critical scenarios. Leveraging distinctive hazardous features, we develop three core risk-aware components. First, we introduce a risk-incorporated scene encoder, which augments conventional encoders with quantitative risk information to achieve risk-aware encoding of hazardous scene contexts. Next, we incorporate endpoint-risk-combined intention queries as prediction priors in the decoder to ensure that the predicted multimodal trajectories cover both various spatial intentions and risk levels. Lastly, an auxiliary risk prediction task is implemented for the ultimate risk-aware prediction. Furthermore, to support model training and performance evaluation, we introduce a safety-critical trajectory prediction dataset and tailored evaluation metrics. We conduct comprehensive evaluations and compare our model with several SOTA models. Results demonstrate the superior performance of our model, with a significant improvement in most metrics. This prediction advancement enables autonomous vehicles to execute correct collision avoidance maneuvers under safety-critical scenarios, eventually enhancing road traffic safety.

Recent advances in implicit scene representation enable high-fidelity street view novel view synthesis. However, existing methods optimize a neural radiance field for each scene, relying heavily on dense training images and extensive computation resources. To mitigate this shortcoming, we introduce a new method called Efficient Depth-Guided Urban View Synthesis (EDUS) for fast feed-forward inference and efficient per-scene fine-tuning. Different from prior generalizable methods that infer geometry based on feature matching, EDUS leverages noisy predicted geometric priors as guidance to enable generalizable urban view synthesis from sparse input images. The geometric priors allow us to apply our generalizable model directly in the 3D space, gaining robustness across various sparsity levels. Through comprehensive experiments on the KITTI-360 and Waymo datasets, we demonstrate promising generalization abilities on novel street scenes. Moreover, our results indicate that EDUS achieves state-of-the-art performance in sparse view settings when combined with fast test-time optimization., Comment: ECCV2024, Project page: https://xdimlab.github.io/EDUS/

Current Facial Action Unit (FAU) detection methods generally encounter difficulties due to the scarcity of labeled video training data and the limited number of training face IDs, which renders the trained feature extractor insufficient coverage for modeling the large diversity of inter-person facial structures and movements. To explicitly address the above challenges, we propose a novel video-level pre-training scheme by fully exploring the multi-label property of FAUs in the video as well as the temporal label consistency. At the heart of our design is a pre-trained video feature extractor based on the video-masked autoencoder together with a fine-tuning network that jointly completes the multi-level video FAUs analysis tasks, \emph{i.e.} integrating both video-level and frame-level FAU detections, thus dramatically expanding the supervision set from sparse FAUs annotations to ALL video frames including masked ones. Moreover, we utilize inter-frame and intra-frame AU pair state matrices as prior knowledge to guide network training instead of traditional Graph Neural Networks, for better temporal supervision. Our approach demonstrates substantial enhancement in performance compared to the existing state-of-the-art methods used in BP4D and DISFA FAUs datasets.

Range-View(RV)-based 3D point cloud segmentation is widely adopted due to its compact data form. However, RV-based methods fall short in providing robust segmentation for the occluded points and suffer from distortion of projected RGB images due to the sparse nature of 3D point clouds. To alleviate these problems, we propose a new LiDAR and Camera Range-view-based 3D point cloud semantic segmentation method (LaCRange). Specifically, a distortion-compensating knowledge distillation (DCKD) strategy is designed to remedy the adverse effect of RV projection of RGB images. Moreover, a context-based feature fusion module is introduced for robust and preservative sensor fusion. Finally, in order to address the limited resolution of RV and its insufficiency of 3D topology, a new point refinement scheme is devised for proper aggregation of features in 2D and augmentation of point features in 3D. We evaluated the proposed method on large-scale autonomous driving datasets \ie SemanticKITTI and nuScenes. In addition to being real-time, the proposed method achieves state-of-the-art results on nuScenes benchmark

Generating 3D vehicle assets from in-the-wild observations is crucial to autonomous driving. Existing image-to-3D methods cannot well address this problem because they learn generation merely from image RGB information without a deeper understanding of in-the-wild vehicles (such as car models, manufacturers, etc.). This leads to their poor zero-shot prediction capability to handle real-world observations with occlusion or tricky viewing angles. To solve this problem, in this work, we propose VQA-Diff, a novel framework that leverages in-the-wild vehicle images to create photorealistic 3D vehicle assets for autonomous driving. VQA-Diff exploits the real-world knowledge inherited from the Large Language Model in the Visual Question Answering (VQA) model for robust zero-shot prediction and the rich image prior knowledge in the Diffusion model for structure and appearance generation. In particular, we utilize a multi-expert Diffusion Models strategy to generate the structure information and employ a subject-driven structure-controlled generation mechanism to model appearance information. As a result, without the necessity to learn from a large-scale image-to-3D vehicle dataset collected from the real world, VQA-Diff still has a robust zero-shot image-to-novel-view generation ability. We conduct experiments on various datasets, including Pascal 3D+, Waymo, and Objaverse, to demonstrate that VQA-Diff outperforms existing state-of-the-art methods both qualitatively and quantitatively.

Realistic scene reconstruction and view synthesis are essential for advancing autonomous driving systems by simulating safety-critical scenarios. 3D Gaussian Splatting excels in real-time rendering and static scene reconstructions but struggles with modeling driving scenarios due to complex backgrounds, dynamic objects, and sparse views. We propose AutoSplat, a framework employing Gaussian splatting to achieve highly realistic reconstructions of autonomous driving scenes. By imposing geometric constraints on Gaussians representing the road and sky regions, our method enables multi-view consistent simulation of challenging scenarios including lane changes. Leveraging 3D templates, we introduce a reflected Gaussian consistency constraint to supervise both the visible and unseen side of foreground objects. Moreover, to model the dynamic appearance of foreground objects, we estimate residual spherical harmonics for each foreground Gaussian. Extensive experiments on Pandaset and KITTI demonstrate that AutoSplat outperforms state-of-the-art methods in scene reconstruction and novel view synthesis across diverse driving scenarios. Visit our project page at https://autosplat.github.io/.

Appearance-based supervised methods with full-face image input have made tremendous advances in recent gaze estimation tasks. However, intensive human annotation requirement inhibits current methods from achieving industrial level accuracy and robustness. Although current unsupervised pre-training frameworks have achieved success in many image recognition tasks, due to the deep coupling between facial and eye features, such frameworks are still deficient in extracting useful gaze features from full-face. To alleviate above limitations, this work proposes a novel unsupervised/self-supervised gaze pre-training framework, which forces the full-face branch to learn a low dimensional gaze embedding without gaze annotations, through collaborative feature contrast and squeeze modules. In the heart of this framework is an alternating eye-attended/unattended masking training scheme, which squeezes gaze-related information from full-face branch into an eye-masked auto-encoder through an injection bottleneck design that successfully encourages the model to pays more attention to gaze direction rather than facial textures only, while still adopting the eye self-reconstruction objective. In the same time, a novel eye/gaze-related information contrastive loss has been designed to further boost the learned representation by forcing the model to focus on eye-centered regions. Extensive experimental results on several gaze benchmarks demonstrate that the proposed scheme achieves superior performances over unsupervised state-of-the-art., Comment: 12 pages, 6 figures, 7 tables

Spatial optimal responses to both inlet disturbances and harmonic external forcing for hypersonic flows over a blunt cone at nonzero angles of attack are obtained by efficiently solving the direct-adjoint equations with a parabolic approach. In either case, the most amplified disturbances initially take the form of localized streamwise vortices on the windward side and will undergo a two-stage evolution process when propagating downstream: they first experience a substantial algebraic growth by exploiting the Orr and lift-up mechanisms, and then smoothly transition to a quasi exponential-growth stage driven by the crossflow-instability mechanism, accompanied by an azimuthal advection of the disturbance structure towards the leeward side. The algebraic-growth phase is most receptive to the external forcing, whereas the exponential-growth stage relies on the disturbance frequency and can be significantly strengthened by increasing the angle of attack. The wavemaker delineating the structural sensitivity region for the optimal gain is shown to lie on the windward side immediately downstream of the inlet, implying a potent control strategy. Additionally, considerable non-modal growth is also observed for broadband high-frequency disturbances residing in the entropy layer.

All-dielectric optical resonators, exhibiting exotic near-field distributions upon excitations, have emerged as low-loss, versatile and highly adaptable components in nanophotonic structures for manipulating electromagnetic waves and enhancing light-matter interactions. However, achieving experimental full three-dimensional characterization of near-fields within dielectric materials poses significant challenges. Here, we develop a novel technique using high-order sideband generation to image near-field wave patterns inside dielectric optical resonators. By exploiting the phase-sensitivity of various harmonic orders that enables the detection of near-field distributions at distinct depths, we realize three-dimensional tomographic and super-resolution near-field imaging inside a micrometer-thick silicon anapole resonator. Furthermore, our method offers high-contrast polarization sensitivity and phase-resolving capability, providing comprehensive vectorial near-field information. Our approach can potentially be applied to diverse dielectric metamaterials, and becomes a valuable tool for comprehensive characterization of near-field wave phenomena within dielectric materials., Comment: 26 pages, 4 figures

Topological flat bands formed in two-dimensional lattice systems offer unique opportunity to study the fractional phases of matter in the absence of an external magnetic field. Celebrated examples include fractional quantum anomalous Hall (FQAH) effects and fractional topological insulators. Recently, FQAH effects have been experimentally realized in both the twisted bilayer MoTe2 (tMoTe2) system and the rhombohedral stacked multilayer graphene/hBN moir\'e systems. To date, experimental studies mainly focus on the first moir\'e flat band, except a very recent work that studied novel transport properties in higher moir\'e bands of a 2.1{\deg} tMoTe2 device. Here, we present the systematical transport study of approximately 3{\deg} tMoTe2 devices, especially for the second moir\'e band. At {\nu} = -2 and -4, time-reversal-symmetric single and double quantum spin Hall states formed, consistent with the previous observation in 2.1{\deg} tMoTe2 device. On the other hand, we observed ferromagnetism in the second moir\'e band, and a Chern insulator state driven by out-of-plane magnetic fields at {\nu} = -3. At {\nu} = -2.2 to -2.7, finite temperature resistivity minimum with 1/T scaling at low temperatures, and large out-of-plane negative magnetoresistance have been observed. Applying out-of-plane electric field can induce quantum phase transitions at both integer and fractional filling factors. Our studies pave the way for realizing tunable topological states and other unexpected magnetic phases beyond the first moir\'e flat band based on twisted MoTe2 platform.

Learning-based intelligent energy management systems for plug-in hybrid electric vehicles (PHEVs) are crucial for achieving efficient energy utilization. However, their application faces system reliability challenges in the real world, which prevents widespread acceptance by original equipment manufacturers (OEMs). This paper begins by establishing a PHEV model based on physical and data-driven models, focusing on the high-fidelity training environment. It then proposes a real-vehicle application-oriented control framework, combining horizon-extended reinforcement learning (RL)-based energy management with the equivalent consumption minimization strategy (ECMS) to enhance practical applicability, and improves the flawed method of equivalent factor evaluation based on instantaneous driving cycle and powertrain states found in existing research. Finally, comprehensive simulation and hardware-in-the-loop validation are carried out which demonstrates the advantages of the proposed control framework in fuel economy over adaptive-ECMS and rule-based strategies. Compared to conventional RL architectures that directly control powertrain components, the proposed control method not only achieves similar optimality but also significantly enhances the disturbance resistance of the energy management system, providing an effective control framework for RL-based energy management strategies aimed at real-vehicle applications by OEMs.

The intrinsic magnetic order, large topological-magnetic gap and rich topological phases make MnBi2Te4 a wonderful platform to study exotic topological quantum states such as axion insulator and Chern insulator. To realize and manipulate these topological phases in a MnBi2Te4 thin film, precise manipulation of the electric field across the film is essential, which requires a dual-gate structure. In this work, we achieve dual-gate tuning of MnBi2Te4 thin films grown with molecular beam epitaxy on SrTiO3(111) substrates by applying the substrate and an AlOx layer as the gate dielectrics of bottom and top gates, respectively. Under magnetic field of 9T and temperature of 20 mK, the Hall and longitudinal resistivities of the films show inversed gate-voltage dependence, for both top- and bottom-gates, signifying the existence of the dissipationless edge state contributed by Chern insulator phase in the ferromagnetic configuration. The maximum of the Hall resistivity only reaches 0.8 h/e2, even with dual-gate tuning, probably due to the high density of bulk carriers introduced by secondary phases. In the antiferromagnetic state under zero magnetic field, the films show normal insulator behavior. The dual-gated MnBi2Te4 thin films lay the foundation for developing devices based on electrically tunable topological quantum states., Comment: 24 pages, 4 figures

Orbital angular momentum (OAM) entangled photon pairs with narrow bandwidths play a crucial role in the interaction of light and quantum states of matter. In this article, we demonstrate an approach for generating OAM entangled photon pairs with a narrow bandwidth by using a single driving beam in a $^{85}$Rb atomic vapor cell. This single driving beam is able to simultaneously couple two atomic transitions and directly generate OAM entangled biphotons by leveraging the OAM conservation law through the spontaneous four-wave mixing (SFWM) process. The photon pairs exhibit a maximum cross-correlation function value of 27.7 and a linewidth of 4 MHz. The OAM entanglement is confirmed through quantum state tomography, revealing a fidelity of 95.7\% and a concurrence of 0.926 when compared to the maximally entangled state. Our scheme is notably simpler than previously proposed schemes and represents the first demonstration of generating subnatural-linewidth entangled photon pairs in hot atomic systems., Comment: 10 pages, 6 figures