Search

While speculative decoding has recently appeared as a promising direction for accelerating the inference of large language models (LLMs), the speedup and scalability are strongly bounded by the token acceptance rate. Prevalent methods usually organize predicted tokens as independent chains or fixed token trees, which fails to generalize to diverse query distributions. In this paper, we propose DySpec, a faster speculative decoding algorithm with a novel dynamic token tree structure. We begin by bridging the draft distribution and acceptance rate from intuitive and empirical clues, and successfully show that the two variables are strongly correlated. Based on this, we employ a greedy strategy to dynamically expand the token tree at run time. Theoretically, we show that our method can achieve optimal results under mild assumptions. Empirically, DySpec yields a higher acceptance rate and speedup than fixed trees. DySpec can drastically improve the throughput and reduce the latency of token generation across various data distribution and model sizes, which significantly outperforms strong competitors, including Specinfer and Sequoia. Under low temperature setting, DySpec can improve the throughput up to 9.1$\times$ and reduce the latency up to 9.4$\times$ on Llama2-70B. Under high temperature setting, DySpec can also improve the throughput up to 6.21$\times$, despite the increasing difficulty of speculating more than one token per step for draft model., Comment: 8 pages, 4 figures

LLMs are typically trained to answer user questions or follow instructions similarly to how human experts respond. However, in the standard alignment framework they lack the basic ability of explicit thinking before answering. Thinking is important for complex questions that require reasoning and planning -- but can be applied to any task. We propose a training method for equipping existing LLMs with such thinking abilities for general instruction following without use of additional human data. We achieve this by an iterative search and optimization procedure that explores the space of possible thought generations, allowing the model to learn how to think without direct supervision. For each instruction, the thought candidates are scored using a judge model to evaluate their responses only, and then optimized via preference optimization. We show that this procedure leads to superior performance on AlpacaEval and Arena-Hard, and shows gains from thinking on non-reasoning categories such as marketing, health and general knowledge, in addition to more traditional reasoning & problem-solving tasks.

With hundreds of thousands of language models available on Huggingface today, efficiently evaluating and utilizing these models across various downstream, tasks has become increasingly critical. Many existing methods repeatedly learn task-specific representations of Large Language Models (LLMs), which leads to inefficiencies in both time and computational resources. To address this, we propose EmbedLLM, a framework designed to learn compact vector representations, of LLMs that facilitate downstream applications involving many models, such as model routing. We introduce an encoder-decoder approach for learning such embeddings, along with a systematic framework to evaluate their effectiveness. Empirical results show that EmbedLLM outperforms prior methods in model routing both in accuracy and latency. Additionally, we demonstrate that our method can forecast a model's performance on multiple benchmarks, without incurring additional inference cost. Extensive probing experiments validate that the learned embeddings capture key model characteristics, e.g. whether the model is specialized for coding tasks, even without being explicitly trained on them. We open source our dataset, code and embedder to facilitate further research and application.

Tactile sensing plays a vital role in enabling robots to perform fine-grained, contact-rich tasks. However, the high dimensionality of tactile data, due to the large coverage on dexterous hands, poses significant challenges for effective tactile feature learning, especially for 3D tactile data, as there are no large standardized datasets and no strong pretrained backbones. To address these challenges, we propose a novel canonical representation that reduces the difficulty of 3D tactile feature learning and further introduces a force-based self-supervised pretraining task to capture both local and net force features, which are crucial for dexterous manipulation. Our method achieves an average success rate of 78% across four fine-grained, contact-rich dexterous manipulation tasks in real-world experiments, demonstrating effectiveness and robustness compared to other methods. Further analysis shows that our method fully utilizes both spatial and force information from 3D tactile data to accomplish the tasks. The videos can be viewed at https://3dtacdex.github.io.

Creating crystal bilayers twisted with respect to each other would lead to large periodic supercell structures, which can support a wide range of novel electron correlated phenomena, where the full understanding is still under debate. Here, we propose a new scheme to realize a nonlinearity-induced dynamical self-organized twisted-bilayer lattice in an atomic Bose-Einstein condensate (BEC). The key idea here is to utilize the nonlinear effect from the intrinsic atomic interactions to couple different layers and induce a dynamical self-organized supercell structure, dramatically distinct from the conventional wisdom to achieve the static twisted-bilayer lattices. To illustrate that, we study the dynamics of a two-component BEC and show that the nonlinear interaction effect naturally emerged in the Gross-Pitaevskii equation of interacting bosonic ultracold atoms can dynamically induce both periodic (commensurable) and aperiodic (incommensurable) moir\'{e} structures. One of the interesting moir\'{e} phenomena, i.e., the flat-band physics, is shown through investigating the dynamics of the wave packet of BEC. Our proposal can be implemented using available state-of-the-art experimental techniques and reveal a profound connection between the nonlinearity and twistronics in cold atom quantum simulators., Comment: 6 pages, 5 figures

Large Language Models (LLMs) are rapidly surpassing human knowledge in many domains. While improving these models traditionally relies on costly human data, recent self-rewarding mechanisms (Yuan et al., 2024) have shown that LLMs can improve by judging their own responses instead of relying on human labelers. However, existing methods have primarily focused on improving model responses rather than judgment capabilities, resulting in rapid saturation during iterative training. To address this issue, we introduce a novel Meta-Rewarding step to the self-improvement process, where the model judges its own judgements and uses that feedback to refine its judgment skills. Surprisingly, this unsupervised approach improves the model's ability to judge {\em and} follow instructions, as demonstrated by a win rate improvement of Llama-3-8B-Instruct from 22.9% to 39.4% on AlpacaEval 2, and 20.6% to 29.1% on Arena-Hard. These results strongly suggest the potential for self-improving models without human supervision.

Nuanced expressiveness, particularly through fine-grained hand and facial expressions, is pivotal for enhancing the realism and vitality of digital human representations. In this work, we focus on investigating the expressiveness of human avatars when learned from monocular RGB video; a setting that introduces new challenges in capturing and animating fine-grained details. To this end, we introduce EVA, a drivable human model that meticulously sculpts fine details based on 3D Gaussians and SMPL-X, an expressive parametric human model. Focused on enhancing expressiveness, our work makes three key contributions. First, we highlight the critical importance of aligning the SMPL-X model with RGB frames for effective avatar learning. Recognizing the limitations of current SMPL-X prediction methods for in-the-wild videos, we introduce a plug-and-play module that significantly ameliorates misalignment issues. Second, we propose a context-aware adaptive density control strategy, which is adaptively adjusting the gradient thresholds to accommodate the varied granularity across body parts. Last but not least, we develop a feedback mechanism that predicts per-pixel confidence to better guide the learning of 3D Gaussians. Extensive experiments on two benchmarks demonstrate the superiority of our framework both quantitatively and qualitatively, especially on the fine-grained hand and facial details. See the project website at \url{https://evahuman.github.io}

Large language models (LLMs) exhibit impressive capabilities across a wide range of tasks, yet the choice of which model to use often involves a trade-off between performance and cost. More powerful models, though effective, come with higher expenses, while less capable models are more cost-effective. To address this dilemma, we propose several efficient router models that dynamically select between a stronger and a weaker LLM during inference, aiming to optimize the balance between cost and response quality. We develop a training framework for these routers leveraging human preference data and data augmentation techniques to enhance performance. Our evaluation on widely-recognized benchmarks shows that our approach significantly reduces costs-by over 2 times in certain cases-without compromising the quality of responses. Interestingly, our router models also demonstrate significant transfer learning capabilities, maintaining their performance even when the strong and weak models are changed at test time. This highlights the potential of these routers to provide a cost-effective yet high-performance solution for deploying LLMs.

The rapid evolution of Large Language Models (LLMs) has outpaced the development of model evaluation, highlighting the need for continuous curation of new, challenging benchmarks. However, manual curation of high-quality, human-aligned benchmarks is expensive and time-consuming. To address this, we introduce BenchBuilder, an automated pipeline that leverages LLMs to curate high-quality, open-ended prompts from large, crowd-sourced datasets, enabling continuous benchmark updates without human in the loop. We apply BenchBuilder to datasets such as Chatbot Arena and WildChat-1M, extracting challenging prompts and utilizing LLM-as-a-Judge for automatic model evaluation. To validate benchmark quality, we propose new metrics to measure a benchmark's alignment with human preferences and ability to separate models. We release Arena-Hard-Auto, a benchmark consisting 500 challenging prompts curated by BenchBuilder. Arena-Hard-Auto provides 3x higher separation of model performances compared to MT-Bench and achieves 98.6% correlation with human preference rankings, all at a cost of $20. Our work sets a new framework for the scalable curation of automated benchmarks from extensive data.

In this paper, we propose the optimal production transport model, which is an extension of the classical optimal transport model. We observe in economics, the production of the factories can always be adjusted within a certain range, while the classical optimal transport does not take this situation into account. Therefore, differing from the classical optimal transport, one of the marginals is allowed to vary within a certain range in our proposed model. To address this, we introduce a multiple relaxation optimal production transport model and propose the generalized alternating Sinkhorn algorithms, inspired by the Sinkhorn algorithm and the double regularization method. By incorporating multiple relaxation variables and multiple regularization terms, the inequality and capacity constraints in the optimal production transport model are naturally satisfied. Alternating iteration algorithms are derived based on the duality of the regularized model. We also provide a theoretical analysis to guarantee the convergence of our proposed algorithms. Numerical results indicate significant advantages in terms of accuracy and efficiency. Furthermore, we apply the optimal production transport model to the coal production and transport problem. Numerical simulation demonstrates that our proposed model can save the production and transport cost by 13.17%., Comment: 28 pages, 4 figures

By equipping the most recent 3D Gaussian Splatting representation with head 3D morphable models (3DMM), existing methods manage to create head avatars with high fidelity. However, most existing methods only reconstruct a head without the body, substantially limiting their application scenarios. We found that naively applying Gaussians to model the clothed chest and shoulders tends to result in blurry reconstruction and noisy floaters under novel poses. This is because of the fundamental limitation of Gaussians and point clouds -- each Gaussian or point can only have a single directional radiance without spatial variance, therefore an unnecessarily large number of them is required to represent complicated spatially varying texture, even for simple geometry. In contrast, we propose to model the body part with a neural texture that consists of coarse and pose-dependent fine colors. To properly render the body texture for each view and pose without accurate geometry nor UV mapping, we optimize another sparse set of Gaussians as anchors that constrain the neural warping field that maps image plane coordinates to the texture space. We demonstrate that Gaussian Head & Shoulders can fit the high-frequency details on the clothed upper body with high fidelity and potentially improve the accuracy and fidelity of the head region. We evaluate our method with casual phone-captured and internet videos and show our method archives superior reconstruction quality and robustness in both self and cross reenactment tasks. To fully utilize the efficient rendering speed of Gaussian splatting, we additionally propose an accelerated inference method of our trained model without Multi-Layer Perceptron (MLP) queries and reach a stable rendering speed of around 130 FPS for any subjects., Comment: Project Page: https://gaussian-head-shoulders.netlify.app/

Controllable text-to-image generation synthesizes visual text and objects in images with certain conditions, which are frequently applied to emoji and poster generation. Visual text rendering and layout-to-image generation tasks have been popular in controllable text-to-image generation. However, each of these tasks typically focuses on single modality generation or rendering, leaving yet-to-be-bridged gaps between the approaches correspondingly designed for each of the tasks. In this paper, we combine text rendering and layout-to-image generation tasks into a single task: layout-controllable text-object synthesis (LTOS) task, aiming at synthesizing images with object and visual text based on predefined object layout and text contents. As compliant datasets are not readily available for our LTOS task, we construct a layout-aware text-object synthesis dataset, containing elaborate well-aligned labels of visual text and object information. Based on the dataset, we propose a layout-controllable text-object adaptive fusion (TOF) framework, which generates images with clear, legible visual text and plausible objects. We construct a visual-text rendering module to synthesize text and employ an object-layout control module to generate objects while integrating the two modules to harmoniously generate and integrate text content and objects in images. To better the image-text integration, we propose a self-adaptive cross-attention fusion module that helps the image generation to attend more to important text information. Within such a fusion module, we use a self-adaptive learnable factor to learn to flexibly control the influence of cross-attention outputs on image generation. Experimental results show that our method outperforms the state-of-the-art in LTOS, text rendering, and layout-to-image tasks, enabling harmonious visual text rendering and object generation.

Driven by the great advances in metaverse and edge computing technologies, vehicular edge metaverses are expected to disrupt the current paradigm of intelligent transportation systems. As highly computerized avatars of Vehicular Metaverse Users (VMUs), the Vehicle Twins (VTs) deployed in edge servers can provide valuable metaverse services to improve driving safety and on-board satisfaction for their VMUs throughout journeys. To maintain uninterrupted metaverse experiences, VTs must be migrated among edge servers following the movements of vehicles. This can raise concerns about privacy breaches during the dynamic communications among vehicular edge metaverses. To address these concerns and safeguard location privacy, pseudonyms as temporary identifiers can be leveraged by both VMUs and VTs to realize anonymous communications in the physical space and virtual spaces. However, existing pseudonym management methods fall short in meeting the extensive pseudonym demands in vehicular edge metaverses, thus dramatically diminishing the performance of privacy preservation. To this end, we present a cross-metaverse empowered dual pseudonym management framework. We utilize cross-chain technology to enhance management efficiency and data security for pseudonyms. Furthermore, we propose a metric to assess the privacy level and employ a Multi-Agent Deep Reinforcement Learning (MADRL) approach to obtain an optimal pseudonym generating strategy. Numerical results demonstrate that our proposed schemes are high-efficiency and cost-effective, showcasing their promising applications in vehicular edge metaverses., Comment: 14 pages, 9 figures

3D decomposition/segmentation still remains a challenge as large-scale 3D annotated data is not readily available. Contemporary approaches typically leverage 2D machine-generated segments, integrating them for 3D consistency. While the majority of these methods are based on NeRFs, they face a potential weakness that the instance/semantic embedding features derive from independent MLPs, thus preventing the segmentation network from learning the geometric details of the objects directly through radiance and density. In this paper, we propose ClusteringSDF, a novel approach to achieve both segmentation and reconstruction in 3D via the neural implicit surface representation, specifically Signal Distance Function (SDF), where the segmentation rendering is directly integrated with the volume rendering of neural implicit surfaces. Although based on ObjectSDF++, ClusteringSDF no longer requires the ground-truth segments for supervision while maintaining the capability of reconstructing individual object surfaces, but purely with the noisy and inconsistent labels from pre-trained models.As the core of ClusteringSDF, we introduce a high-efficient clustering mechanism for lifting the 2D labels to 3D and the experimental results on the challenging scenes from ScanNet and Replica datasets show that ClusteringSDF can achieve competitive performance compared against the state-of-the-art with significantly reduced training time., Comment: Project Page: https://sm0kywu.github.io/ClusteringSDF/

In real-world scenarios, objects often require repositioning and reorientation before they can be grasped, a process known as pre-grasp manipulation. Learning universal dexterous functional pre-grasp manipulation requires precise control over the relative position, orientation, and contact between the hand and object while generalizing to diverse dynamic scenarios with varying objects and goal poses. To address this challenge, we propose a teacher-student learning approach that utilizes a novel mutual reward, incentivizing agents to optimize three key criteria jointly. Additionally, we introduce a pipeline that employs a mixture-of-experts strategy to learn diverse manipulation policies, followed by a diffusion policy to capture complex action distributions from these experts. Our method achieves a success rate of 72.6\% across more than 30 object categories by leveraging extrinsic dexterity and adjusting from feedback.

We study a millicharged co-interacting dark matter scenario, where the primary dark matter constituent is the dark photon $A'$ and the secondary component is the fermion $\chi$. In this model, $\chi$ interacts with $A'$ via a $U(1)'$ interaction while being millicharged with respect to normal photons. Our investigation focuses on the oscillation of $A'$ dark matter into photons within the background of $\chi$ particles, revealing that the $A'-\chi$ scattering rate benefits from a Bose enhancement of the $A'$ final state. As the oscillation production rate is directly linked to the scattering rate, the conversion of $A'$ dark matter into monochromatic photons experiences significant amplification owing to this Bose enhancement, especially when the scattering rate $\Gamma_{\rm sca}$ approaches the dark photon mass $m_{A'}$. These converted monochromatic photons are detectable through radio telescopes and can induce distortions in the Cosmic Microwave Background (CMB) spectrum. We find that the sensitivity of radio telescopes and the constraints imposed by CMB distortion on the kinetic mixing parameter are notably heightened compared to scenarios without the subdominant millicharged dark matter., Comment: 9 pages, 1 figure

Finding cohesive subgraphs in a large graph has many important applications, such as community detection and biological network analysis. Clique is often a too strict cohesive structure since communities or biological modules rarely form as cliques for various reasons such as data noise. Therefore, $k$-plex is introduced as a popular clique relaxation, which is a graph where every vertex is adjacent to all but at most $k$ vertices. In this paper, we propose a fast branch-and-bound algorithm as well as its task-based parallel version to enumerate all maximal $k$-plexes with at least $q$ vertices. Our algorithm adopts an effective search space partitioning approach that provides a lower time complexity, a new pivot vertex selection method that reduces candidate vertex size, an effective upper-bounding technique to prune useless branches, and three novel pruning techniques by vertex pairs. Our parallel algorithm uses a timeout mechanism to eliminate straggler tasks, and maximizes cache locality while ensuring load balancing. Extensive experiments show that compared with the state-of-the-art algorithms, our sequential and parallel algorithms enumerate large maximal $k$-plexes with up to $5 \times$ and $18.9 \times$ speedup, respectively. Ablation results also demonstrate that our pruning techniques bring up to $7 \times$ speedup compared with our basic algorithm., Comment: Accepted by EDBT2025. Camera-ready version

With the continuous advancement in autonomous systems, it becomes crucial to provide robust safety guarantees for safety-critical systems. Hamilton-Jacobi Reachability Analysis is a formal verification method that guarantees performance and safety for dynamical systems and is widely applicable to various tasks and challenges. Traditionally, reachability problems are solved by using grid-based methods, whose computational and memory cost scales exponentially with the dimensionality of the system. To overcome this challenge, DeepReach, a deep learning-based approach that approximately solves high-dimensional reachability problems, is proposed and has shown lots of promise. In this paper, we aim to improve the performance of DeepReach on high-dimensional systems by exploring different choices of activation functions. We first run experiments on a 3D system as a proof of concept. Then we demonstrate the effectiveness of our approach on a 9D multi-vehicle collision problem.

Knowledge distillation is one of the methods for model compression, and existing knowledge distillation techniques focus on how to improve the distillation algorithm so as to enhance the distillation efficdiency. This paper introduces dynamic incremental learning into knowledge distillation and proposes a distillation strategy for education distillation. Specifically, it is proposed to look at fragmented student models divided from the full student model as low models. As the grade level rises, fragmented student models deepen in conjunction with designed teaching reference layers, while learning and distilling from more teacher models. By moving from lower to higher grades, fragmented student models were gradually integrated into a complete target student model, and the performance of the student models gradually improved from lower to senior grades of the stage. Education distillation strategies combined with distillation algorithms outperform the results of single distillation algorithms on the public dataset CIFAR100,Caltech256, Food-101 dataset.

Large Language Models (LLMs) can acquire extensive world knowledge through pre-training on large corpora. However, due to exposure to low-quality data, LLMs may exhibit harmful behavior without aligning with human values. The dominant approach for steering LLMs towards beneficial behavior involves Reinforcement Learning with Human Feedback (RLHF), with Proximal Policy Optimization (PPO) serving as the default RL optimizer. Despite its effectiveness, PPO has limitations when optimizing rewards trained from comparison-based loss. Primarily, PPO is not invariant to equivalent reward functions containing identical preference information due to the need to calibrate the reward scale. Additionally, PPO's necessity for token-wise updates introduces complexity in both function approximation and algorithm design compared to trajectory-wise optimization. This paper proposes a new framework, reinforcement learning with relative feedback, and a novel trajectory-wise policy gradient algorithm, Pairwise Proximal Policy Optimization (P3O) that operates directly on comparative rewards. We show theoretically that P3O is invariant to equivalent rewards and avoids the complexity of PPO. Empirical evaluations demonstrate that P3O outperforms PPO in the KL-Reward trade-off and can align with human preferences as well as or better than prior methods. In summary, this work introduces a simpler yet effective approach for aligning LLMs to human preferences through relative feedback., Comment: 19 pages, 5 figures

PyPose is an open-source library for robot learning. It combines a learning-based approach with physics-based optimization, which enables seamless end-to-end robot learning. It has been used in many tasks due to its meticulously designed application programming interface (API) and efficient implementation. From its initial launch in early 2022, PyPose has experienced significant enhancements, incorporating a wide variety of new features into its platform. To satisfy the growing demand for understanding and utilizing the library and reduce the learning curve of new users, we present the fundamental design principle of the imperative programming interface, and showcase the flexible usage of diverse functionalities and modules using an extremely simple Dubins car example. We also demonstrate that the PyPose can be easily used to navigate a real quadruped robot with a few lines of code.

In this paper, we present a comprehensive investigation into the sensitivity of a Rydberg atom electric field sensor, with a specific focus on the minimum detectable field (MDF) as a key metric. The study utilizes one-mode Floquet theory to calculate the Stark shift for selected Rydberg states when exposed to a signal electric field. The results are compared to those obtained using the rotating wave approximation (RWA). To enhance the sensor's sensitivity when the frequency of the signal electric field deviates from resonance frequencies between Rydberg states, we propose incorporating an extra coupling electric field and using many-mode Floquet theory, a generalization of one-mode Floquet theory, to theoretically analyze this kind of Rydberg atom electric field sensor. The Autler-Townes effect resulting from this coupling electric field causes Rydberg states to split into dressed states, effectively increasing sensitivity by modulating the frequencies of resonance peaks. Moreover, the phenomenon of two-photon absorption in the presence of the coupling electric field is explored. We demonstrate that by appropriately adjusting the coupling electric field's amplitude or frequency, one can control the occurrence of two-photon resonances, providing additional sensitivity enhancement for the Rydberg sensor within the significantly extended off-resonance domain. The study underscores the significance of coupling fields in enhancing the sensitivity of Rydberg atom electric field sensors. These insights hold promising implications for the development of more robust and versatile electric field sensing devices, applicable in diverse fields such as precision measurements and quantum information processing.

The use of anthropomorphic robotic hands for assisting individuals in situations where human hands may be unavailable or unsuitable has gained significant importance. In this paper, we propose a novel task called human-assisting dexterous grasping that aims to train a policy for controlling a robotic hand's fingers to assist users in grasping objects. Unlike conventional dexterous grasping, this task presents a more complex challenge as the policy needs to adapt to diverse user intentions, in addition to the object's geometry. We address this challenge by proposing an approach consisting of two sub-modules: a hand-object-conditional grasping primitive called Grasping Gradient Field~(GraspGF), and a history-conditional residual policy. GraspGF learns `how' to grasp by estimating the gradient from a success grasping example set, while the residual policy determines `when' and at what speed the grasping action should be executed based on the trajectory history. Experimental results demonstrate the superiority of our proposed method compared to baselines, highlighting the user-awareness and practicality in real-world applications. The codes and demonstrations can be viewed at "https://sites.google.com/view/graspgf".

We investigate the enhancement of dark sector particle production within the scattering dominant regime. These particles typically exhibit a slight mixing with Standard Model particles through various portals, allowing for their generation through in-medium oscillation from Standard Model particle sources. Our analysis reveals that in the scattering dominance regime, with a significantly smaller scattering mean free path $\lambda_{\rm sca}$ compared to the absorption mean free path $\lambda_{\rm abs}$, the nonresonant production of sterile states can experience an enhancement by a factor of $\lambda_{\rm abs}/\lambda_{\rm sca}$. This phenomenon is demonstrated within the context of kinetic mixing dark photon production at a reactor, precisely satisfying this condition. By incorporating this collisional enhancement, we find that the current sensitivity to the mixing parameter $\epsilon$ for dark photons in the TEXONO experiment can be significantly improved across a range spanning from tens of eV to MeV. This advancement establishes the most stringent laboratory constraint within this mass spectrum for the dark photon. Sterile neutrino production, however, does not exhibit such enhancement, either due to the failure to meet the scattering dominance criterion or the neutrino damping in resonant production., Comment: 8 pages, 4 figures, PRD accepted

The radiance fields style transfer is an emerging field that has recently gained popularity as a means of 3D scene stylization, thanks to the outstanding performance of neural radiance fields in 3D reconstruction and view synthesis. We highlight a research gap in radiance fields style transfer, the lack of sufficient perceptual controllability, motivated by the existing concept in the 2D image style transfer. In this paper, we present ARF-Plus, a 3D neural style transfer framework offering manageable control over perceptual factors, to systematically explore the perceptual controllability in 3D scene stylization. Four distinct types of controls - color preservation control, (style pattern) scale control, spatial (selective stylization area) control, and depth enhancement control - are proposed and integrated into this framework. Results from real-world datasets, both quantitative and qualitative, show that the four types of controls in our ARF-Plus framework successfully accomplish their corresponding perceptual controls when stylizing 3D scenes. These techniques work well for individual style inputs as well as for the simultaneous application of multiple styles within a scene. This unlocks a realm of limitless possibilities, allowing customized modifications of stylization effects and flexible merging of the strengths of different styles, ultimately enabling the creation of novel and eye-catching stylistic effects on 3D scenes.

Given the revolutionary role of metaverses, healthcare metaverses are emerging as a transformative force, creating intelligent healthcare systems that offer immersive and personalized services. The healthcare metaverses allow for effective decision-making and data analytics for users. However, there still exist critical challenges in building healthcare metaverses, such as the risk of sensitive data leakage and issues with sensing data security and freshness, as well as concerns around incentivizing data sharing. In this paper, we first design a user-centric privacy-preserving framework based on decentralized Federated Learning (FL) for healthcare metaverses. To further improve the privacy protection of healthcare metaverses, a cross-chain empowered FL framework is utilized to enhance sensing data security. This framework utilizes a hierarchical cross-chain architecture with a main chain and multiple subchains to perform decentralized, privacy-preserving, and secure data training in both virtual and physical spaces. Moreover, we utilize Age of Information (AoI) as an effective data-freshness metric and propose an AoI-based contract theory model under Prospect Theory (PT) to motivate sensing data sharing in a user-centric manner. This model exploits PT to better capture the subjective utility of the service provider. Finally, our numerical results demonstrate the effectiveness of the proposed schemes for healthcare metaverses.

Simultaneous localization and mapping (SLAM) is a fundamental task for numerous applications such as autonomous navigation and exploration. Despite many SLAM datasets have been released, current SLAM solutions still struggle to have sustained and resilient performance. One major issue is the absence of high-quality datasets including diverse all-weather conditions and a reliable metric for assessing robustness. This limitation significantly restricts the scalability and generalizability of SLAM technologies, impacting their development, validation, and deployment. To address this problem, we present SubT-MRS, an extremely challenging real-world dataset designed to push SLAM towards all-weather environments to pursue the most robust SLAM performance. It contains multi-degraded environments including over 30 diverse scenes such as structureless corridors, varying lighting conditions, and perceptual obscurants like smoke and dust; multimodal sensors such as LiDAR, fisheye camera, IMU, and thermal camera; and multiple locomotions like aerial, legged, and wheeled robots. We develop accuracy and robustness evaluation tracks for SLAM and introduced novel robustness metrics. Comprehensive studies are performed, revealing new observations, challenges, and opportunities for future research.

Generating complete 360-degree panoramas from narrow field of view images is ongoing research as omnidirectional RGB data is not readily available. Existing GAN-based approaches face some barriers to achieving higher quality output, and have poor generalization performance over different mask types. In this paper, we present our 360-degree indoor RGB-D panorama outpainting model using latent diffusion models (LDM), called PanoDiffusion. We introduce a new bi-modal latent diffusion structure that utilizes both RGB and depth panoramic data during training, which works surprisingly well to outpaint depth-free RGB images during inference. We further propose a novel technique of introducing progressive camera rotations during each diffusion denoising step, which leads to substantial improvement in achieving panorama wraparound consistency. Results show that our PanoDiffusion not only significantly outperforms state-of-the-art methods on RGB-D panorama outpainting by producing diverse well-structured results for different types of masks, but can also synthesize high-quality depth panoramas to provide realistic 3D indoor models., Comment: Project Page: https://sm0kywu.github.io/panodiffusion/

Multi armed bandit (MAB) algorithms have been increasingly used to complement or integrate with A/B tests and randomized clinical trials in e-commerce, healthcare, and policymaking. Recent developments incorporate possible delayed feedback. While existing MAB literature often focuses on maximizing the expected cumulative reward outcomes (or, equivalently, regret minimization), few efforts have been devoted to establish valid statistical inference approaches to quantify the uncertainty of learned policies. We attempt to fill this gap by providing a unified statistical inference framework for policy evaluation where a target policy is allowed to differ from the data collecting policy, and our framework allows delay to be associated with the treatment arms. We present an adaptively weighted estimator that on one hand incorporates the arm-dependent delaying mechanism to achieve consistency, and on the other hand mitigates the variance inflation across stages due to vanishing sampling probability. In particular, our estimator does not critically depend on the ability to estimate the unknown delay mechanism. Under appropriate conditions, we prove that our estimator converges to a normal distribution as the number of time points goes to infinity, which provides guarantees for large-sample statistical inference. We illustrate the finite-sample performance of our approach through Monte Carlo experiments., Comment: 25 pages, 8 figures, ICML 2023

While deep learning techniques have become extremely popular for solving a broad range of optimization problems, methods to enforce hard constraints during optimization, particularly on deep neural networks, remain underdeveloped. Inspired by the rich literature on meshless interpolation and its extension to spectral collocation methods in scientific computing, we develop a series of approaches for enforcing hard constraints on neural fields, which we refer to as Constrained Neural Fields (CNF). The constraints can be specified as a linear operator applied to the neural field and its derivatives. We also design specific model representations and training strategies for problems where standard models may encounter difficulties, such as conditioning of the system, memory consumption, and capacity of the network when being constrained. Our approaches are demonstrated in a wide range of real-world applications. Additionally, we develop a framework that enables highly efficient model and constraint specification, which can be readily applied to any downstream task where hard constraints need to be explicitly satisfied during optimization., Comment: 37th Conference on Neural Information Processing Systems (NeurIPS 2023)

Neural view synthesis (NVS) is one of the most successful techniques for synthesizing free viewpoint videos, capable of achieving high fidelity from only a sparse set of captured images. This success has led to many variants of the techniques, each evaluated on a set of test views typically using image quality metrics such as PSNR, SSIM, or LPIPS. There has been a lack of research on how NVS methods perform with respect to perceived video quality. We present the first study on perceptual evaluation of NVS and NeRF variants. For this study, we collected two datasets of scenes captured in a controlled lab environment as well as in-the-wild. In contrast to existing datasets, these scenes come with reference video sequences, allowing us to test for temporal artifacts and subtle distortions that are easily overlooked when viewing only static images. We measured the quality of videos synthesized by several NVS methods in a well-controlled perceptual quality assessment experiment as well as with many existing state-of-the-art image/video quality metrics. We present a detailed analysis of the results and recommendations for dataset and metric selection for NVS evaluation.

Implicit surface representations such as the signed distance function (SDF) have emerged as a promising approach for image-based surface reconstruction. However, existing optimization methods assume solid surfaces and are therefore unable to properly reconstruct semi-transparent surfaces and thin structures, which also exhibit low opacity due to the blending effect with the background. While neural radiance field (NeRF) based methods can model semi-transparency and achieve photo-realistic quality in synthesized novel views, their volumetric geometry representation tightly couples geometry and opacity, and therefore cannot be easily converted into surfaces without introducing artifacts. We present $\alpha$Surf, a novel surface representation with decoupled geometry and opacity for the reconstruction of semi-transparent and thin surfaces where the colors mix. Ray-surface intersections on our representation can be found in closed-form via analytical solutions of cubic polynomials, avoiding Monte-Carlo sampling and is fully differentiable by construction. Our qualitative and quantitative evaluations show that our approach can accurately reconstruct surfaces with semi-transparent and thin parts with fewer artifacts, achieving better reconstruction quality than state-of-the-art SDF and NeRF methods. Website: https://alphasurf.netlify.app/

We study stochastic delayed feedback in general multi-agent sequential decision making, which includes bandits, single-agent Markov decision processes (MDPs), and Markov games (MGs). We propose a novel reduction-based framework, which turns any multi-batched algorithm for sequential decision making with instantaneous feedback into a sample-efficient algorithm that can handle stochastic delays in sequential decision making. By plugging different multi-batched algorithms into our framework, we provide several examples demonstrating that our framework not only matches or improves existing results for bandits, tabular MDPs, and tabular MGs, but also provides the first line of studies on delays in sequential decision making with function approximation. In summary, we provide a complete set of sharp results for multi-agent sequential decision making with delayed feedback., Comment: Accepted by Neurips 2023. arXiv admin note: text overlap with arXiv:2110.14555 by other authors

Capacity constrained optimal transport is a variant of optimal transport, which adds extra constraints on the set of feasible couplings in the original optimal transport problem to limit the mass transported between each pair of source and sink. Based on this setting, constrained optimal transport has numerous applications, e.g., finance, network flow. However, due to the large number of constraints in this problem, existing algorithms are both time-consuming and space-consuming. In this paper, inspired by entropic regularization for the classical optimal transport problem, we introduce a novel regularization term for capacity constrained optimal transport. The regularized problem naturally satisfies the capacity constraints and consequently makes it possible to analyze the duality. Unlike the matrix-vector multiplication in the alternate iteration scheme for solving classical optimal transport, in our algorithm, each alternate iteration step is to solve several single-variable equations. Fortunately, we find that each of these equations corresponds to a single-variable monotonic function, and we convert solving these equations into finding the unique zero point of each single-variable monotonic function with Newton's method. Extensive numerical experiments demonstrate that our proposed method has a significant advantage in terms of accuracy, efficiency, and memory consumption compared with existing methods., Comment: 17 pages, 3 figures

Deep learning has had remarkable success in robotic perception, but its data-centric nature suffers when it comes to generalizing to ever-changing environments. By contrast, physics-based optimization generalizes better, but it does not perform as well in complicated tasks due to the lack of high-level semantic information and reliance on manual parametric tuning. To take advantage of these two complementary worlds, we present PyPose: a robotics-oriented, PyTorch-based library that combines deep perceptual models with physics-based optimization. PyPose's architecture is tidy and well-organized, it has an imperative style interface and is efficient and user-friendly, making it easy to integrate into real-world robotic applications. Besides, it supports parallel computing of any order gradients of Lie groups and Lie algebras and $2^{\text{nd}}$-order optimizers, such as trust region methods. Experiments show that PyPose achieves more than $10\times$ speedup in computation compared to the state-of-the-art libraries. To boost future research, we provide concrete examples for several fields of robot learning, including SLAM, planning, control, and inertial navigation., Comment: Project Website: https://pypose.org Documentation: https://pypose.org/docs/ Tutorial: https://pypose.org/tutorials/ Source code: https://github.com/pypose/pypose

Space-based infrared tiny ship detection aims at separating tiny ships from the images captured by earth orbiting satellites. Due to the extremely large image coverage area (e.g., thousands square kilometers), candidate targets in these images are much smaller, dimer, more changeable than those targets observed by aerial-based and land-based imaging devices. Existing short imaging distance-based infrared datasets and target detection methods cannot be well adopted to the space-based surveillance task. To address these problems, we develop a space-based infrared tiny ship detection dataset (namely, NUDT-SIRST-Sea) with 48 space-based infrared images and 17598 pixel-level tiny ship annotations. Each image covers about 10000 square kilometers of area with 10000X10000 pixels. Considering the extreme characteristics (e.g., small, dim, changeable) of those tiny ships in such challenging scenes, we propose a multi-level TransUNet (MTU-Net) in this paper. Specifically, we design a Vision Transformer (ViT) Convolutional Neural Network (CNN) hybrid encoder to extract multi-level features. Local feature maps are first extracted by several convolution layers and then fed into the multi-level feature extraction module (MVTM) to capture long-distance dependency. We further propose a copy-rotate-resize-paste (CRRP) data augmentation approach to accelerate the training phase, which effectively alleviates the issue of sample imbalance between targets and background. Besides, we design a FocalIoU loss to achieve both target localization and shape description. Experimental results on the NUDT-SIRST-Sea dataset show that our MTU-Net outperforms traditional and existing deep learning based SIRST methods in terms of probability of detection, false alarm rate and intersection over union.

We discuss the Wormholes in general dimensions by studying the Einstein-phantom scalar field with and without the cosmological constant. Solving AdS wormholes in general dimension is hard due to the nonlinear nature of the theory. In this work, we implement the AdS/Ricci-flat correspondence, extended to include the axion field (the phantom scalar field), to construct AdS wormholes. Wormholes of Ellis-Bronnikov class are discussed in general dimensions., Comment: Corrected some typos in the Appendix in previous version

Commercial depth sensors usually generate noisy and missing depths, especially on specular and transparent objects, which poses critical issues to downstream depth or point cloud-based tasks. To mitigate this problem, we propose a powerful RGBD fusion network, SwinDRNet, for depth restoration. We further propose Domain Randomization-Enhanced Depth Simulation (DREDS) approach to simulate an active stereo depth system using physically based rendering and generate a large-scale synthetic dataset that contains 130K photorealistic RGB images along with their simulated depths carrying realistic sensor noises. To evaluate depth restoration methods, we also curate a real-world dataset, namely STD, that captures 30 cluttered scenes composed of 50 objects with different materials from specular, transparent, to diffuse. Experiments demonstrate that the proposed DREDS dataset bridges the sim-to-real domain gap such that, trained on DREDS, our SwinDRNet can seamlessly generalize to other real depth datasets, e.g. ClearGrasp, and outperform the competing methods on depth restoration with a real-time speed. We further show that our depth restoration effectively boosts the performance of downstream tasks, including category-level pose estimation and grasping tasks. Our data and code are available at https://github.com/PKU-EPIC/DREDS, Comment: ECCV 2022

Achieving human-level dexterity is an important open problem in robotics. However, tasks of dexterous hand manipulation, even at the baby level, are challenging to solve through reinforcement learning (RL). The difficulty lies in the high degrees of freedom and the required cooperation among heterogeneous agents (e.g., joints of fingers). In this study, we propose the Bimanual Dexterous Hands Benchmark (Bi-DexHands), a simulator that involves two dexterous hands with tens of bimanual manipulation tasks and thousands of target objects. Specifically, tasks in Bi-DexHands are designed to match different levels of human motor skills according to cognitive science literature. We built Bi-DexHands in the Issac Gym; this enables highly efficient RL training, reaching 30,000+ FPS by only one single NVIDIA RTX 3090. We provide a comprehensive benchmark for popular RL algorithms under different settings; this includes Single-agent/Multi-agent RL, Offline RL, Multi-task RL, and Meta RL. Our results show that the PPO type of on-policy algorithms can master simple manipulation tasks that are equivalent up to 48-month human babies (e.g., catching a flying object, opening a bottle), while multi-agent RL can further help to master manipulations that require skilled bimanual cooperation (e.g., lifting a pot, stacking blocks). Despite the success on each single task, when it comes to acquiring multiple manipulation skills, existing RL algorithms fail to work in most of the multi-task and the few-shot learning settings, which calls for more substantial development from the RL community. Our project is open sourced at https://github.com/PKU-MARL/DexterousHands., Comment: 38 pages, 8 figures

Given a monocular video, segmenting and decoupling dynamic objects while recovering the static environment is a widely studied problem in machine intelligence. Existing solutions usually approach this problem in the image domain, limiting their performance and understanding of the environment. We introduce Decoupled Dynamic Neural Radiance Field (D$^2$NeRF), a self-supervised approach that takes a monocular video and learns a 3D scene representation which decouples moving objects, including their shadows, from the static background. Our method represents the moving objects and the static background by two separate neural radiance fields with only one allowing for temporal changes. A naive implementation of this approach leads to the dynamic component taking over the static one as the representation of the former is inherently more general and prone to overfitting. To this end, we propose a novel loss to promote correct separation of phenomena. We further propose a shadow field network to detect and decouple dynamically moving shadows. We introduce a new dataset containing various dynamic objects and shadows and demonstrate that our method can achieve better performance than state-of-the-art approaches in decoupling dynamic and static 3D objects, occlusion and shadow removal, and image segmentation for moving objects.

Data is the driving force of machine learning, with the amount and quality of training data often being more important for the performance of a system than architecture and training details. But collecting, processing and annotating real data at scale is difficult, expensive, and frequently raises additional privacy, fairness and legal concerns. Synthetic data is a powerful tool with the potential to address these shortcomings: 1) it is cheap 2) supports rich ground-truth annotations 3) offers full control over data and 4) can circumvent or mitigate problems regarding bias, privacy and licensing. Unfortunately, software tools for effective data generation are less mature than those for architecture design and training, which leads to fragmented generation efforts. To address these problems we introduce Kubric, an open-source Python framework that interfaces with PyBullet and Blender to generate photo-realistic scenes, with rich annotations, and seamlessly scales to large jobs distributed over thousands of machines, and generating TBs of data. We demonstrate the effectiveness of Kubric by presenting a series of 13 different generated datasets for tasks ranging from studying 3D NeRF models to optical flow estimation. We release Kubric, the used assets, all of the generation code, as well as the rendered datasets for reuse and modification., Comment: 21 pages, CVPR2022

Grasping moving objects, such as goods on a belt or living animals, is an important but challenging task in robotics. Conventional approaches rely on a set of manually defined object motion patterns for training, resulting in poor generalization to unseen object trajectories. In this work, we introduce an adversarial reinforcement learning framework for dynamic grasping, namely GraspARL. To be specific. we formulate the dynamic grasping problem as a 'move-and-grasp' game, where the robot is to pick up the object on the mover and the adversarial mover is to find a path to escape it. Hence, the two agents play a min-max game and are trained by reinforcement learning. In this way, the mover can auto-generate diverse moving trajectories while training. And the robot trained with the adversarial trajectories can generalize to various motion patterns. Empirical results on the simulator and real-world scenario demonstrate the effectiveness of each and good generalization of our method.

Smart Internet of Vehicles (IoVs) combined with Artificial Intelligence (AI) will contribute to vehicle decision-making in the Intelligent Transportation System (ITS). Multi-Vehicle Pursuit games (MVP), a multi-vehicle cooperative ability to capture mobile targets, is becoming a hot research topic gradually. Although there are some achievements in the field of MVP in the open space environment, the urban area brings complicated road structures and restricted moving spaces as challenges to the resolution of MVP games. We define an Observation-constrained MVP (OMVP) problem in this paper and propose a Transformer-based Time and Team Reinforcement Learning scheme ($ \text{T}^3 $OMVP) to address the problem. First, a new multi-vehicle pursuit model is constructed based on decentralized partially observed Markov decision processes (Dec-POMDP) to instantiate this problem. Second, by introducing and modifying the transformer-based observation sequence, QMIX is redefined to adapt to the complicated road structure, restricted moving spaces and constrained observations, so as to control vehicles to pursue the target combining the vehicle's observations. Third, a multi-intersection urban environment is built to verify the proposed scheme. Extensive experimental results demonstrate that the proposed $ \text{T}^3 $OMVP scheme achieves significant improvements relative to state-of-the-art QMIX approaches by 9.66%~106.25%. Code is available at https://github.com/pipihaiziguai/T3OMVP.