Search

Collaborative perception has garnered considerable attention due to its capacity to address several inherent challenges in single-agent perception, including occlusion and out-of-range issues. However, existing collaborative perception systems heavily rely on precise localization systems to establish a consistent spatial coordinate system between agents. This reliance makes them susceptible to large pose errors or malicious attacks, resulting in substantial reductions in perception performance. To address this, we propose~$\mathtt{CoBEVGlue}$, a novel self-localized collaborative perception system, which achieves more holistic and robust collaboration without using an external localization system. The core of~$\mathtt{CoBEVGlue}$ is a novel spatial alignment module, which provides the relative poses between agents by effectively matching co-visible objects across agents. We validate our method on both real-world and simulated datasets. The results show that i) $\mathtt{CoBEVGlue}$ achieves state-of-the-art detection performance under arbitrary localization noises and attacks; and ii) the spatial alignment module can seamlessly integrate with a majority of previous methods, enhancing their performance by an average of $57.7\%$. Code is available at https://github.com/VincentNi0107/CoBEVGlue

Offline Black-Box Optimization (BBO) aims at optimizing a black-box function using the knowledge from a pre-collected offline dataset of function values and corresponding input designs. However, the high-dimensional and highly-multimodal input design space of black-box function pose inherent challenges for most existing methods that model and operate directly upon input designs. These issues include but are not limited to high sample complexity, which relates to inaccurate approximation of black-box function; and insufficient coverage and exploration of input design modes, which leads to suboptimal proposal of new input designs. In this work, we consider finding a latent space that serves as a compressed yet accurate representation of the design-value joint space, enabling effective latent exploration of high-value input design modes. To this end, we formulate an learnable energy-based latent space, and propose Noise-intensified Telescoping density-Ratio Estimation (NTRE) scheme for variational learning of an accurate latent space model without costly Markov Chain Monte Carlo. The optimization process is then exploration of high-value designs guided by the learned energy-based model in the latent space, formulated as gradient-based sampling from a latent-variable-parameterized inverse model. We show that our particular parameterization encourages expanded exploration around high-value design modes, motivated by inversion thinking of a fundamental result of conditional covariance matrix typically used for variance reduction. We observe that our method, backed by an accurately learned informative latent space and an expanding-exploration model design, yields significant improvements over strong previous methods on both synthetic and real world datasets such as the design-bench suite.

We introduce Integer Scale, a novel post-training quantization scheme for large language models that effectively resolves the inference bottleneck in current fine-grained quantization approaches while maintaining similar accuracies. Integer Scale is a free lunch as it requires no extra calibration or fine-tuning which will otherwise incur additional costs. It can be used plug-and-play for most fine-grained quantization methods. Its integration results in at most 1.85x end-to-end speed boost over the original counterpart with comparable accuracy. Additionally, due to the orchestration of the proposed Integer Scale and fine-grained quantization, we resolved the quantization difficulty for Mixtral-8x7B and LLaMA-3 models with negligible performance degradation, and it comes with an end-to-end speed boost of 2.13x, and 2.31x compared with their FP16 versions respectively.

Vehicle-to-everything-aided autonomous driving (V2X-AD) has a huge potential to provide a safer driving solution. Despite extensive researches in transportation and communication to support V2X-AD, the actual utilization of these infrastructures and communication resources in enhancing driving performances remains largely unexplored. This highlights the necessity of collaborative autonomous driving: a machine learning approach that optimizes the information sharing strategy to improve the driving performance of each vehicle. This effort necessitates two key foundations: a platform capable of generating data to facilitate the training and testing of V2X-AD, and a comprehensive system that integrates full driving-related functionalities with mechanisms for information sharing. From the platform perspective, we present V2Xverse, a comprehensive simulation platform for collaborative autonomous driving. This platform provides a complete pipeline for collaborative driving. From the system perspective, we introduce CoDriving, a novel end-to-end collaborative driving system that properly integrates V2X communication over the entire autonomous pipeline, promoting driving with shared perceptual information. The core idea is a novel driving-oriented communication strategy. Leveraging this strategy, CoDriving improves driving performance while optimizing communication efficiency. We make comprehensive benchmarks with V2Xverse, analyzing both modular performance and closed-loop driving performance. Experimental results show that CoDriving: i) significantly improves the driving score by 62.49% and drastically reduces the pedestrian collision rate by 53.50% compared to the SOTA end-to-end driving method, and ii) achieves sustaining driving performance superiority over dynamic constraint communication conditions.

Scene simulation in autonomous driving has gained significant attention because of its huge potential for generating customized data. However, existing editable scene simulation approaches face limitations in terms of user interaction efficiency, multi-camera photo-realistic rendering and external digital assets integration. To address these challenges, this paper introduces ChatSim, the first system that enables editable photo-realistic 3D driving scene simulations via natural language commands with external digital assets. To enable editing with high command flexibility,~ChatSim leverages a large language model (LLM) agent collaboration framework. To generate photo-realistic outcomes, ChatSim employs a novel multi-camera neural radiance field method. Furthermore, to unleash the potential of extensive high-quality digital assets, ChatSim employs a novel multi-camera lighting estimation method to achieve scene-consistent assets' rendering. Our experiments on Waymo Open Dataset demonstrate that ChatSim can handle complex language commands and generate corresponding photo-realistic scene videos., Comment: CVPR 2024(Highlight)

Collaborative perception aims to mitigate the limitations of single-agent perception, such as occlusions, by facilitating data exchange among multiple agents. However, most current works consider a homogeneous scenario where all agents use identity sensors and perception models. In reality, heterogeneous agent types may continually emerge and inevitably face a domain gap when collaborating with existing agents. In this paper, we introduce a new open heterogeneous problem: how to accommodate continually emerging new heterogeneous agent types into collaborative perception, while ensuring high perception performance and low integration cost? To address this problem, we propose HEterogeneous ALliance (HEAL), a novel extensible collaborative perception framework. HEAL first establishes a unified feature space with initial agents via a novel multi-scale foreground-aware Pyramid Fusion network. When heterogeneous new agents emerge with previously unseen modalities or models, we align them to the established unified space with an innovative backward alignment. This step only involves individual training on the new agent type, thus presenting extremely low training costs and high extensibility. To enrich agents' data heterogeneity, we bring OPV2V-H, a new large-scale dataset with more diverse sensor types. Extensive experiments on OPV2V-H and DAIR-V2X datasets show that HEAL surpasses SOTA methods in performance while reducing the training parameters by 91.5% when integrating 3 new agent types. We further implement a comprehensive codebase at: https://github.com/yifanlu0227/HEAL, Comment: Accepted by ICLR 2024. The code and data are open-sourced at https://github.com/yifanlu0227/HEAL

Diverse video captioning aims to generate a set of sentences to describe the given video in various aspects. Mainstream methods are trained with independent pairs of a video and a caption from its ground-truth set without exploiting the intra-set relationship, resulting in low diversity of generated captions. Different from them, we formulate diverse captioning into a semantic-concept-guided set prediction (SCG-SP) problem by fitting the predicted caption set to the ground-truth set, where the set-level relationship is fully captured. Specifically, our set prediction consists of two synergistic tasks, i.e., caption generation and an auxiliary task of concept combination prediction providing extra semantic supervision. Each caption in the set is attached to a concept combination indicating the primary semantic content of the caption and facilitating element alignment in set prediction. Furthermore, we apply a diversity regularization term on concepts to encourage the model to generate semantically diverse captions with various concept combinations. These two tasks share multiple semantics-specific encodings as input, which are obtained by iterative interaction between visual features and conceptual queries. The correspondence between the generated captions and specific concept combinations further guarantees the interpretability of our model. Extensive experiments on benchmark datasets show that the proposed SCG-SP achieves state-of-the-art (SOTA) performance under both relevance and diversity metrics., Comment: aaai 2024 accepted

Recent observations of the high-redshift universe have uncovered a significant number of active galactic nuclei, implying that supermassive black holes (SMBHs) would have to have been formed at much earlier times than expected. Direct collapse of metal-free gas clouds to SMBHs after recombination could help explain the early formation of SMBHs, but this scenario is stymied by the fragmentation of the clouds due to efficient molecular hydrogen cooling. We show that a subdominant population of tiny, evaporating primordial black holes, with significant clustering in some gas clouds, can heat the gas sufficiently so that molecular hydrogen is not formed, and direct collapse to to black holes is possible even at high redshifts., Comment: 15 pages, 4 figures

The large language model era urges faster and less costly inference. Prior model compression works on LLMs tend to undertake a software-centric approach primarily focused on the simulated quantization performance. By neglecting the feasibility of deployment, these approaches are typically disabled in real practice. They used to drastically push down the quantization bit range for a reduced computation which might not be supported by the mainstream hardware, or involve sophisticated algorithms that introduce extra computation or memory access overhead. We argue that pursuing a hardware-centric approach in the construction of quantization algorithms is crucial. In this regard, we are driven to build our compression method on top of hardware awareness, eliminating impractical algorithm choices while maximizing the benefit of hardware acceleration. Our method, OdysseyLLM, comes with a novel W4A8 kernel implementation called FastGEMM and a combined recipe of quantization strategies. Extensive experiments manifest the superiority of our W4A8 method which brings the actual speed boosting up to \textbf{4$\times$} compared to Hugging Face FP16 inference and \textbf{2.23$\times$} vs. the state-of-the-art inference engine TensorRT-LLM in FP16, and \textbf{1.45$\times$} vs. TensorRT-LLM in INT8, yet without substantially harming the performance.

Collaborative perception can substantially boost each agent's perception ability by facilitating communication among multiple agents. However, temporal asynchrony among agents is inevitable in the real world due to communication delays, interruptions, and clock misalignments. This issue causes information mismatch during multi-agent fusion, seriously shaking the foundation of collaboration. To address this issue, we propose CoBEVFlow, an asynchrony-robust collaborative perception system based on bird's eye view (BEV) flow. The key intuition of CoBEVFlow is to compensate motions to align asynchronous collaboration messages sent by multiple agents. To model the motion in a scene, we propose BEV flow, which is a collection of the motion vector corresponding to each spatial location. Based on BEV flow, asynchronous perceptual features can be reassigned to appropriate positions, mitigating the impact of asynchrony. CoBEVFlow has two advantages: (i) CoBEVFlow can handle asynchronous collaboration messages sent at irregular, continuous time stamps without discretization; and (ii) with BEV flow, CoBEVFlow only transports the original perceptual features, instead of generating new perceptual features, avoiding additional noises. To validate CoBEVFlow's efficacy, we create IRregular V2V(IRV2V), the first synthetic collaborative perception dataset with various temporal asynchronies that simulate different real-world scenarios. Extensive experiments conducted on both IRV2V and the real-world dataset DAIR-V2X show that CoBEVFlow consistently outperforms other baselines and is robust in extremely asynchronous settings. The code is available at https://github.com/MediaBrain-SJTU/CoBEVFlow., Comment: 16 pages, 9 figures. Accepted by NeurIPS 2023

To protect the intellectual property of well-trained deep neural networks (DNNs), black-box watermarks, which are embedded into the prediction behavior of DNN models on a set of specially-crafted samples and extracted from suspect models using only API access, have gained increasing popularity in both academy and industry. Watermark robustness is usually implemented against attackers who steal the protected model and obfuscate its parameters for watermark removal. However, current robustness evaluations are primarily performed under moderate attacks or unrealistic settings. Existing removal attacks could only crack a small subset of the mainstream black-box watermarks, and fall short in four key aspects: incomplete removal, reliance on prior knowledge of the watermark, performance degradation, and high dependency on data. In this paper, we propose a watermark-agnostic removal attack called \textsc{Neural Dehydration} (\textit{abbrev.} \textsc{Dehydra}), which effectively erases all ten mainstream black-box watermarks from DNNs, with only limited or even no data dependence. In general, our attack pipeline exploits the internals of the protected model to recover and unlearn the watermark message. We further design target class detection and recovered sample splitting algorithms to reduce the utility loss and achieve data-free watermark removal on five of the watermarking schemes. We conduct comprehensive evaluation of \textsc{Dehydra} against ten mainstream black-box watermarks on three benchmark datasets and DNN architectures. Compared with existing removal attacks, \textsc{Dehydra} achieves strong removal effectiveness across all the covered watermarks, preserving at least $90\%$ of the stolen model utility, under the data-limited settings, i.e., less than $2\%$ of the training data or even data-free., Comment: 20 pages, ver 2.0, accepted by CCS '24

In the presence of (relatively) long-range forces, structures can form even during the radiation dominated era, leading to compact objects, such as Fermi balls or primordial black holes (PBHs), which can account for all or part of dark matter. We present a detailed analysis of a model in which fermions are produced from the inflaton decay developing some particle-antiparticle asymmetry. These fermions undergo clustering and structure formation driven by a Yukawa interaction. The same interaction provides a cooling channel for the dark halos via scalar radiation, leading to rapid collapse and the formation of a compact object. We discuss the criteria for the formation of either PBHs and Fermi balls. In the PBH formation regime, supermassive PBHs can seed the active galactic nuclei or quasars found at high redshift. Alternatively, Fermi balls can account for all of the cold dark matter, while evading microlensing constraints., Comment: 9 pages, 6 figures

Virtual Reality (VR), Augmented Reality (AR), Mixed Reality (MR), digital twin, Metaverse and other related digital technologies have attracted much attention in recent years. These new emerging technologies are changing the world significantly. This research introduces a fusion model, i.e. Fusion Universe (FU), where the virtual, physical, and cognitive worlds are merged together. Therefore, it is crucial to establish a set of principles for the fusion model that is compatible with our physical universe laws and principles. This paper investigates several aspects that could affect immersive and interactive experience; and proposes the fundamental principles for Fusion Universe that can integrate physical and virtual world seamlessly., Comment: 5 pages, 8 figures

Camera-only 3D detection provides an economical solution with a simple configuration for localizing objects in 3D space compared to LiDAR-based detection systems. However, a major challenge lies in precise depth estimation due to the lack of direct 3D measurements in the input. Many previous methods attempt to improve depth estimation through network designs, e.g., deformable layers and larger receptive fields. This work proposes an orthogonal direction, improving the camera-only 3D detection by introducing multi-agent collaborations. Our proposed collaborative camera-only 3D detection (CoCa3D) enables agents to share complementary information with each other through communication. Meanwhile, we optimize communication efficiency by selecting the most informative cues. The shared messages from multiple viewpoints disambiguate the single-agent estimated depth and complement the occluded and long-range regions in the single-agent view. We evaluate CoCa3D in one real-world dataset and two new simulation datasets. Results show that CoCa3D improves previous SOTA performances by 44.21% on DAIR-V2X, 30.60% on OPV2V+, 12.59% on CoPerception-UAVs+ for AP@70. Our preliminary results show a potential that with sufficient collaboration, the camera might overtake LiDAR in some practical scenarios. We released the dataset and code at https://siheng-chen.github.io/dataset/CoPerception+ and https://github.com/MediaBrain-SJTU/CoCa3D., Comment: Accepted by CVPR23

Automated driving systems rely on 3D object detectors to recognize possible obstacles from LiDAR point clouds. However, recent works show the adversary can forge non-existent cars in the prediction results with a few fake points (i.e., appearing attack). By removing statistical outliers, existing defenses are however designed for specific attacks or biased by predefined heuristic rules. Towards more comprehensive mitigation, we first systematically inspect the mechanism of recent appearing attacks: Their common weaknesses are observed in crafting fake obstacles which (i) have obvious differences in the local parts compared with real obstacles and (ii) violate the physical relation between depth and point density. In this paper, we propose a novel plug-and-play defensive module which works by side of a trained LiDAR-based object detector to eliminate forged obstacles where a major proportion of local parts have low objectness, i.e., to what degree it belongs to a real object. At the core of our module is a local objectness predictor, which explicitly incorporates the depth information to model the relation between depth and point density, and predicts each local part of an obstacle with an objectness score. Extensive experiments show, our proposed defense eliminates at least 70% cars forged by three known appearing attacks in most cases, while, for the best previous defense, less than 30% forged cars are eliminated. Meanwhile, under the same circumstance, our defense incurs less overhead for AP/precision on cars compared with existing defenses. Furthermore, We validate the effectiveness of our proposed defense on simulation-based closed-loop control driving tests in the open-source system of Baidu's Apollo., Comment: Accepted by USENIX Sercurity 2023

Collaborative 3D object detection exploits information exchange among multiple agents to enhance accuracy of object detection in presence of sensor impairments such as occlusion. However, in practice, pose estimation errors due to imperfect localization would cause spatial message misalignment and significantly reduce the performance of collaboration. To alleviate adverse impacts of pose errors, we propose CoAlign, a novel hybrid collaboration framework that is robust to unknown pose errors. The proposed solution relies on a novel agent-object pose graph modeling to enhance pose consistency among collaborating agents. Furthermore, we adopt a multi-scale data fusion strategy to aggregate intermediate features at multiple spatial resolutions. Comparing with previous works, which require ground-truth pose for training supervision, our proposed CoAlign is more practical since it doesn't require any ground-truth pose supervision in the training and makes no specific assumptions on pose errors. Extensive evaluation of the proposed method is carried out on multiple datasets, certifying that CoAlign significantly reduce relative localization error and achieving the state of art detection performance when pose errors exist. Code are made available for the use of the research community at https://github.com/yifanlu0227/CoAlign., Comment: Accepted by ICRA2023. This version updates the final results with data augmentation and correct AP calculations(sort the detections of all samples by confidence)

We propose a novel domain adaptive action detection approach and a new adaptation protocol that leverages the recent advancements in image-level unsupervised domain adaptation (UDA) techniques and handle vagaries of instance-level video data. Self-training combined with cross-domain mixed sampling has shown remarkable performance gain in semantic segmentation in UDA (unsupervised domain adaptation) context. Motivated by this fact, we propose an approach for human action detection in videos that transfers knowledge from the source domain (annotated dataset) to the target domain (unannotated dataset) using mixed sampling and pseudo-label-based selftraining. The existing UDA techniques follow a ClassMix algorithm for semantic segmentation. However, simply adopting ClassMix for action detection does not work, mainly because these are two entirely different problems, i.e., pixel-label classification vs. instance-label detection. To tackle this, we propose a novel action instance mixed sampling technique that combines information across domains based on action instances instead of action classes. Moreover, we propose a new UDA training protocol that addresses the long-tail sample distribution and domain shift problem by using supervision from an auxiliary source domain (ASD). For the ASD, we propose a new action detection dataset with dense frame-level annotations. We name our proposed framework as domain-adaptive action instance mixing (DA-AIM). We demonstrate that DA-AIM consistently outperforms prior works on challenging domain adaptation benchmarks. The source code is available at https://github.com/wwwfan628/DA-AIM.

The task of Dense Video Captioning (DVC) aims to generate captions with timestamps for multiple events in one video. Semantic information plays an important role for both localization and description of DVC. We present a semantic-assisted dense video captioning model based on the encoding-decoding framework. In the encoding stage, we design a concept detector to extract semantic information, which is then fused with multi-modal visual features to sufficiently represent the input video. In the decoding stage, we design a classification head, paralleled with the localization and captioning heads, to provide semantic supervision. Our method achieves significant improvements on the YouMakeup dataset under DVC evaluation metrics and achieves high performance in the Makeup Dense Video Captioning (MDVC) task of PIC 4th Challenge., Comment: 5 pages, 2 figures, report of PIC 4th Challenge Accepted by PIC'22 workshop

Graph Neural Networks (GNNs) are the first choice for learning algorithms on graph data. GNNs promise to integrate (i) node features as well as (ii) edge information in an end-to-end learning algorithm. How does this promise work out practically? In this paper, we study to what extend GNNs are necessary to solve prominent graph classification problems. We find that for graph classification, a GNN is not more than the sum of its parts. We also find that, unlike features, predictions with an edge-only model do not always transfer to GNNs.

AI-for-science approaches have been applied to solve scientific problems (e.g., nuclear fusion, ecology, genomics, meteorology) and have achieved highly promising results. Spatial precipitation downscaling is one of the most important meteorological problem and urgently requires the participation of AI. However, the lack of a well-organized and annotated large-scale dataset hinders the training and verification of more effective and advancing deep-learning models for precipitation downscaling. To alleviate these obstacles, we present the first large-scale spatial precipitation downscaling dataset named RainNet, which contains more than $62,400$ pairs of high-quality low/high-resolution precipitation maps for over $17$ years, ready to help the evolution of deep learning models in precipitation downscaling. Specifically, the precipitation maps carefully collected in RainNet cover various meteorological phenomena (e.g., hurricane, squall), which is of great help to improve the model generalization ability. In addition, the map pairs in RainNet are organized in the form of image sequences ($720$ maps per month or 1 map/hour), showing complex physical properties, e.g., temporal misalignment, temporal sparse, and fluid properties. Furthermore, two deep-learning-oriented metrics are specifically introduced to evaluate or verify the comprehensive performance of the trained model (e.g., prediction maps reconstruction accuracy). To illustrate the applications of RainNet, 14 state-of-the-art models, including deep models and traditional approaches, are evaluated. To fully explore potential downscaling solutions, we propose an implicit physical estimation benchmark framework to learn the above characteristics. Extensive experiments demonstrate the value of RainNet in training and evaluating downscaling models. Our dataset is available at https://neuralchen.github.io/RainNet/., Comment: Accepted at NeurIPS 2022. Project page: https://neuralchen.github.io/RainNet/

We numerically investigate the sphere partition function of a Chern-Simons-matter theory with $SU(N)$ gauge group at level $k$ coupled to three adjoint chiral multiplets that is dual to massive IIA theory. Beyond the leading order $N^{5/3}$ behavior of the free energy, we find numerical evidence for a term of the form $(2/9)\log N-(1/18)\log k$. We conjecture that the $(2/9)\log N$ term may be universal in theories with $N^{5/3}$ scaling in the large-$N$ limit with the Chern-Simons level $k$ held fixed., Comment: 17 pages, 3 figures, minor typos corrected

Micro-vibration has been a dominant factor impairing the performance of scientific experiments which are expected to be deployed in a micro-gravity environment such as Spacelab. The micro-vibration has a serious impact on scientific experiments requiring a quasi-static environment. Therefore, we proposed a maglev vibration isolation platform (MVIP) operating in six degrees of freedom (DOF) to fulfill the environmental requirements. In view of non-contact and large stroke requirements for micro-vibration isolation, an optimization method was utilized to design the actuator. Mathematical models of the actuator's remarkable nonlinearity were established so that its output can be compensated according to floater's varying position and the system's performance may be satisfied. Furthermore, aiming to adapt to an energy-limited environment such as Spacelab, an optimum allocation scheme was put forward. Considering the actuator's nonlinearity, accuracy and minimum energy-consumption can be obtained simultaneously. In view of operating in six DOF, methods for nonlinear compensation and system decoupling were discussed, the necessary controller was also presented. Simulation and experiments validate the system's performance. With a movement range of 10x10x8 mm and rotations of 200 mrad, the decay ratio of -40 dB/Dec between 1-10 Hz was obtained under closed-loop control., Comment: A preprint version

Modern neural network modules which can significantly enhance the learning power usually add too much computational complexity to the original neural networks. In this paper, we pursue very efficient neural network modules which can significantly boost the learning power of deep convolutional neural networks with negligible extra computational cost. We first present both theoretically and experimentally that elementwise square operator has a potential to enhance the learning power of neural networks. Then, we design four types of lightweight modules with elementwise square operators, named as Square-Pooling, Square-Softmin, Square-Excitation, and Square-Encoding. We add our four lightweight modules to Resnet18, Resnet50, and ShuffleNetV2 for better performance in the experiment on ImageNet 2012 dataset. The experimental results show that our modules can bring significant accuracy improvements to the base convolutional neural network models. The performance of our lightweight modules is even comparable to many complicated modules such as bilinear pooling, Squeeze-and-Excitation, and Gather-Excite. Our highly efficient modules are particularly suitable for mobile models. For example, when equipped with a single Square-Pooling module, the top-1 classification accuracy of ShuffleNetV2-0.5x on ImageNet 2012 is absolutely improved by 1.45% with no additional parameters and negligible inference time overhead., Comment: We has submitted this paper to a conference before March 22, 2019. We will improve this paper according to the received reviews. The code will be released if the paper is accepted for publication

We present the first practical attack on a hardware AES accelerator with 256 bit embedded keys using DFA. We identify the challenges of adapting well-known theoretical AES DFA models to hardware under attack from voltage fault injection and present solutions to those challenges. As a result, we managed to recover 278 real-world AES-256 keys from a secure computing system in a matter of hours with minimal cost.

We show how voltage glitching can cause timing violations in CMOS behavior. Then we attack a real, security hardened, consumer device to gain code execution and dump the secure boot ROM.