Search

Automatic Heuristic Design (AHD) is an active research area due to its utility in solving complex search and NP-hard combinatorial optimization problems in the real world. The recent advancements in Large Language Models (LLMs) introduce new possibilities by coupling LLMs with evolutionary computation to automatically generate heuristics, known as LLM-based Evolutionary Program Search (LLM-EPS). While previous LLM-EPS studies obtained great performance on various tasks, there is still a gap in understanding the properties of heuristic search spaces and achieving a balance between exploration and exploitation, which is a critical factor in large heuristic search spaces. In this study, we address this gap by proposing two diversity measurement metrics and perform an analysis on previous LLM-EPS approaches, including FunSearch, EoH, and ReEvo. Results on black-box AHD problems reveal that while EoH demonstrates higher diversity than FunSearch and ReEvo, its objective score is unstable. Conversely, ReEvo's reflection mechanism yields good objective scores but fails to optimize diversity effectively. With this finding in mind, we introduce HSEvo, an adaptive LLM-EPS framework that maintains a balance between diversity and convergence with a harmony search algorithm. Through experimentation, we find that HSEvo achieved high diversity indices and good objective scores while remaining cost-effective. These results underscore the importance of balancing exploration and exploitation and understanding heuristic search spaces in designing frameworks in LLM-EPS., Comment: 18 pages, 12 figures

Motion forecasting (MF) for autonomous driving aims at anticipating trajectories of surrounding agents in complex urban scenarios. In this work, we investigate a mixed strategy in MF training that first pre-train motion forecasters on pseudo-labeled data, then fine-tune them on annotated data. To obtain pseudo-labeled trajectories, we propose a simple pipeline that leverages off-the-shelf single-frame 3D object detectors and non-learning trackers. The whole pre-training strategy including pseudo-labeling is coined as PPT. Our extensive experiments demonstrate that: (1) combining PPT with supervised fine-tuning on annotated data achieves superior performance on diverse testbeds, especially under annotation-efficient regimes, (2) scaling up to multiple datasets improves the previous state-of-the-art and (3) PPT helps enhance cross-dataset generalization. Our findings showcase PPT as a promising pre-training solution for robust motion forecasting in diverse autonomous driving contexts.

Quantum computing has emerged as a powerful tool for solving complex computational problems, but access to real quantum hardware remains limited due to high costs and increasing demand for efficient quantum simulations. Unfortunately, software simulators on CPUs/GPUs such as Qiskit, ProjectQ, and Qsun offer flexibility and support for a large number of qubits, they struggle with high power consumption and limited processing speed, especially as qubit counts scale. Accordingly, quantum emulators implemented on dedicated hardware, such as FPGAs and analog circuits, offer a promising path for addressing energy efficiency concerns. However, existing studies on hardware-based emulators still face challenges in terms of limited flexibility, lack of fidelity evaluation, and power consumption. To overcome these gaps, we propose FQsun, a quantum emulator that enhances performance by integrating four key innovations: efficient memory organization, a configurable Quantum Gate Unit (QGU), optimized scheduling, and multiple number precisions. Five FQsun versions with different number precisions, including 16-bit floating point, 32-bit floating point, 16-bit fixed point, 24-bit fixed point, and 32-bit fixed point, are implemented on the Xilinx ZCU102 FPGA, utilizing between 9,226 and 18,093 LUTs, 1,440 and 7,031 FFs, 344 and 464 BRAMs, and 14 and 88 DSPs and consuming a maximum power of 2.41W. Experimental results demonstrate high accuracy in normalized gate speed, fidelity, and mean square error, particularly with 32-bit fixed-point and floating-point versions, establishing FQsun's capability as a precise quantum emulator. Benchmarking on quantum algorithms such as Quantum Fourier Transform, Parameter-Shift Rule, and Random Quantum Circuits reveals that FQsun achieves superior power-delay product, outperforming traditional software simulators on powerful CPUs by up to 9,870 times., Comment: 17 pages, 14 figures, submitted to the IEEE Transaction on Quantum Engineering

Domain adaptation has been extensively investigated in computer vision but still requires access to target data at the training time, which might be difficult to obtain in some uncommon conditions. In this paper, we present a new framework for domain adaptation relying on a single Vision-Language (VL) latent embedding instead of full target data. First, leveraging a contrastive language-image pre-training model (CLIP), we propose prompt/photo-driven instance normalization (PIN). PIN is a feature augmentation method that mines multiple visual styles using a single target VL latent embedding, by optimizing affine transformations of low-level source features. The VL embedding can come from a language prompt describing the target domain, a partially optimized language prompt, or a single unlabeled target image. Second, we show that these mined styles (i.e., augmentations) can be used for zero-shot (i.e., target-free) and one-shot unsupervised domain adaptation. Experiments on semantic segmentation demonstrate the effectiveness of the proposed method, which outperforms relevant baselines in the zero-shot and one-shot settings., Comment: Under review

Quantum emulators play an important role in the development and testing of quantum algorithms, especially given the limitations of the current FTQC era. Developing high-speed, memory-optimized quantum emulators is a growing research trend, with gate fusion being a promising technique. However, existing gate fusion implementations often struggle to efficiently support large-scale quantum systems with a high number of qubits due to a lack of optimizations for the exponential growth in memory requirements. Therefore, this study proposes the EMMS (Efficient-Memory Matrix Storage) method for storing quantum operators and states, along with an EMMS-based Quantum Emulator Accelerator (QEA) architecture that incorporates multiple processing elements (PEs) to accelerate tensor product and matrix multiplication computations in quantum emulation with gate fusion. The theoretical analysis of the QEA on the Xilinx ZCU102 FPGA, using varying numbers of PEs and different depths of unitary and local data memory, reveals a linear increase in memory depth with the number of qubits. This scaling highlights the potential of the EMMS-based QEA to accommodate larger quantum circuits, providing insights into selecting appropriate memory sizes and FPGA devices. Furthermore, the estimated performance of the QEA with PE counts ranging from $2^2$ to $2^5$ on the Xilinx ZCU102 FPGA demonstrates that increasing the number of PEs significantly reduces the computation cycle count for circuits with fewer than 18 qubits, making it significantly faster than previous works.

We consider the problem of adapting a contrastively pretrained vision-language model like CLIP (Radford et al., 2021) for few-shot classification. The literature addresses this problem by learning a linear classifier of the frozen visual features, optimizing word embeddings, or learning external feature adapters. This paper introduces an alternative way for CLIP adaptation without adding 'external' parameters to optimize. We find that simply fine-tuning the last projection matrix of the vision encoder leads to performance better than all baselines. Furthermore, we show that regularizing training with the distance between the fine-tuned and pretrained matrices adds reliability for adapting CLIP. This simple approach, coined ProLIP, yields state-of-the-art performance on 11 few-shot classification benchmarks, few-shot domain generalization, cross-dataset transfer, base-to-new class generalization, and test-time adaptation. Code will be made available at: https://github.com/astra-vision/ProLIP .

We propose the unified BRAVO challenge to benchmark the reliability of semantic segmentation models under realistic perturbations and unknown out-of-distribution (OOD) scenarios. We define two categories of reliability: (1) semantic reliability, which reflects the model's accuracy and calibration when exposed to various perturbations; and (2) OOD reliability, which measures the model's ability to detect object classes that are unknown during training. The challenge attracted nearly 100 submissions from international teams representing notable research institutions. The results reveal interesting insights into the importance of large-scale pre-training and minimal architectural design in developing robust and reliable semantic segmentation models., Comment: ECCV 2024 proceeding paper of the BRAVO challenge 2024, see https://benchmarks.elsa-ai.eu/?ch=1&com=introduction Corrected numbers in Tables 1,3,4,5 and 10

Diffusion models (DMs) are advanced deep learning models that achieved state-of-the-art capability on a wide range of generative tasks. However, recent studies have shown their vulnerability regarding backdoor attacks, in which backdoored DMs consistently generate a designated result (e.g., a harmful image) called backdoor target when the models' input contains a backdoor trigger. Although various backdoor techniques have been investigated to attack DMs, defense methods against these threats are still limited and underexplored, especially in inverting the backdoor trigger. In this paper, we introduce PureDiffusion, a novel backdoor defense framework that can efficiently detect backdoor attacks by inverting backdoor triggers embedded in DMs. Our extensive experiments on various trigger-target pairs show that PureDiffusion outperforms existing defense methods with a large gap in terms of fidelity (i.e., how much the inverted trigger resembles the original trigger) and backdoor success rate (i.e., the rate that the inverted trigger leads to the corresponding backdoor target). Notably, in certain cases, backdoor triggers inverted by PureDiffusion even achieve higher attack success rate than the original triggers.

We tackle the challenging problem of source-free unsupervised domain adaptation (SFUDA) for 3D semantic segmentation. It amounts to performing domain adaptation on an unlabeled target domain without any access to source data; the available information is a model trained to achieve good performance on the source domain. A common issue with existing SFUDA approaches is that performance degrades after some training time, which is a by product of an under-constrained and ill-posed problem. We discuss two strategies to alleviate this issue. First, we propose a sensible way to regularize the learning problem. Second, we introduce a novel criterion based on agreement with a reference model. It is used (1) to stop the training when appropriate and (2) as validator to select hyperparameters without any knowledge on the target domain. Our contributions are easy to implement and readily amenable for all SFUDA methods, ensuring stable improvements over all baselines. We validate our findings on various 3D lidar settings, achieving state-of-the-art performance. The project repository (with code) is: github.com/valeoai/TTYD., Comment: Accepted to ECCV 2024. Project repository: github.com/valeoai/TTYD

Diffusion models (DMs) have achieved state-of-the-art performance on various generative tasks such as image synthesis, text-to-image, and text-guided image-to-image generation. However, the more powerful the DMs, the more harmful they potentially are. Recent studies have shown that DMs are prone to a wide range of attacks, including adversarial attacks, membership inference, backdoor injection, and various multi-modal threats. Since numerous pre-trained DMs are published widely on the Internet, potential threats from these attacks are especially detrimental to the society, making DM-related security a worth investigating topic. Therefore, in this paper, we conduct a comprehensive survey on the security aspect of DMs, focusing on various attack and defense methods for DMs. First, we present crucial knowledge of DMs with five main types of DMs, including denoising diffusion probabilistic models, denoising diffusion implicit models, noise conditioned score networks, stochastic differential equations, and multi-modal conditional DMs. We further survey a variety of recent studies investigating different types of attacks that exploit the vulnerabilities of DMs. Then, we thoroughly review potential countermeasures to mitigate each of the presented threats. Finally, we discuss open challenges of DM-related security and envision certain research directions for this topic.

Kolmogorov-Arnold Networks (KANs), a novel type of neural network, have recently gained popularity and attention due to the ability to substitute multi-layer perceptions (MLPs) in artificial intelligence (AI) with higher accuracy and interoperability. However, KAN assessment is still limited and cannot provide an in-depth analysis of a specific domain. Furthermore, no study has been conducted on the implementation of KANs in hardware design, which would directly demonstrate whether KANs are truly superior to MLPs in practical applications. As a result, in this paper, we focus on verifying KANs for classification issues, which are a common but significant topic in AI using four different types of datasets. Furthermore, the corresponding hardware implementation is considered using the Vitis high-level synthesis (HLS) tool. To the best of our knowledge, this is the first article to implement hardware for KAN. The results indicate that KANs cannot achieve more accuracy than MLPs in high complex datasets while utilizing substantially higher hardware resources. Therefore, MLP remains an effective approach for achieving accuracy and efficiency in software and hardware implementation., Comment: 6 pages, 3 figures, 2 tables

Quantum compilation is the process of converting a target unitary operation into a trainable unitary represented by a quantum circuit. It has a wide range of applications, including gate optimization, quantum-assisted compiling, quantum state preparation, and quantum dynamic simulation. Traditional quantum compilation usually optimizes circuits for a single target. However, many quantum systems require simultaneous optimization of multiple targets, such as thermal state preparation, time-dependent dynamic simulation, and others. To address this, we develop a multi-target quantum compilation algorithm to improve the performance and flexibility of simulating multiple quantum systems. Our benchmarks and case studies demonstrate the effectiveness of the algorithm, highlighting the importance of multi-target optimization in advancing quantum computing. This work lays the groundwork for further development and evaluation of multi-target quantum compilation algorithms., Comment: 13 pages, 5 figures

Motion forecasting is crucial in autonomous driving systems to anticipate the future trajectories of surrounding agents such as pedestrians, vehicles, and traffic signals. In end-to-end forecasting, the model must jointly detect and track from sensor data (cameras or LiDARs) the past trajectories of the different elements of the scene and predict their future locations. We depart from the current trend of tackling this task via end-to-end training from perception to forecasting, and instead use a modular approach. We individually build and train detection, tracking and forecasting modules. We then only use consecutive finetuning steps to integrate the modules better and alleviate compounding errors. We conduct an in-depth study on the finetuning strategies and it reveals that our simple yet effective approach significantly improves performance on the end-to-end forecasting benchmark. Consequently, our solution ranks first in the Argoverse 2 End-to-end Forecasting Challenge, with 63.82 mAPf. We surpass forecasting results by +17.1 points over last year's winner and by +13.3 points over this year's runner-up. This remarkable performance in forecasting can be explained by our modular paradigm, which integrates finetuning strategies and significantly outperforms the end-to-end-trained counterparts. The code, model weights and results are made available https://github.com/valeoai/valeo4cast., Comment: Winning solution of the Argoverse 2 "Unified Detection, Tracking, and Forecasting" challenge; work accepted at Road++ ECCVW 2024

Vision-Large-Language-Models (Vision-LLMs) are increasingly being integrated into autonomous driving (AD) systems due to their advanced visual-language reasoning capabilities, targeting the perception, prediction, planning, and control mechanisms. However, Vision-LLMs have demonstrated susceptibilities against various types of adversarial attacks, which would compromise their reliability and safety. To further explore the risk in AD systems and the transferability of practical threats, we propose to leverage typographic attacks against AD systems relying on the decision-making capabilities of Vision-LLMs. Different from the few existing works developing general datasets of typographic attacks, this paper focuses on realistic traffic scenarios where these attacks can be deployed, on their potential effects on the decision-making autonomy, and on the practical ways in which these attacks can be physically presented. To achieve the above goals, we first propose a dataset-agnostic framework for automatically generating false answers that can mislead Vision-LLMs' reasoning. Then, we present a linguistic augmentation scheme that facilitates attacks at image-level and region-level reasoning, and we extend it with attack patterns against multiple reasoning tasks simultaneously. Based on these, we conduct a study on how these attacks can be realized in physical traffic scenarios. Through our empirical study, we evaluate the effectiveness, transferability, and realizability of typographic attacks in traffic scenes. Our findings demonstrate particular harmfulness of the typographic attacks against existing Vision-LLMs (e.g., LLaVA, Qwen-VL, VILA, and Imp), thereby raising community awareness of vulnerabilities when incorporating such models into AD systems. We will release our source code upon acceptance., Comment: 12 pages, 5 tables, 5 figures, work in progress

Creating large-scale virtual urban scenes with variant styles is inherently challenging. To facilitate prototypes of virtual production and bypass the need for complex materials and lighting setups, we introduce the first vision-and-text-driven texture stylization system for large-scale urban scenes, StyleCity. Taking an image and text as references, StyleCity stylizes a 3D textured mesh of a large-scale urban scene in a semantics-aware fashion and generates a harmonic omnidirectional sky background. To achieve that, we propose to stylize a neural texture field by transferring 2D vision-and-text priors to 3D globally and locally. During 3D stylization, we progressively scale the planned training views of the input 3D scene at different levels in order to preserve high-quality scene content. We then optimize the scene style globally by adapting the scale of the style image with the scale of the training views. Moreover, we enhance local semantics consistency by the semantics-aware style loss which is crucial for photo-realistic stylization. Besides texture stylization, we further adopt a generative diffusion model to synthesize a style-consistent omnidirectional sky image, which offers a more immersive atmosphere and assists the semantic stylization process. The stylized neural texture field can be baked into an arbitrary-resolution texture, enabling seamless integration into conventional rendering pipelines and significantly easing the virtual production prototyping process. Extensive experiments demonstrate our stylized scenes' superiority in qualitative and quantitative performance and user preferences., Comment: Accepted by ECCV2024. Project page: https://chenyingshu.github.io/stylecity3d/

Referring image segmentation is a challenging task that involves generating pixel-wise segmentation masks based on natural language descriptions. The complexity of this task increases with the intricacy of the sentences provided. Existing methods have relied mostly on visual features to generate the segmentation masks while treating text features as supporting components. However, this under-utilization of text understanding limits the model's capability to fully comprehend the given expressions. In this work, we propose a novel framework that specifically emphasizes object and context comprehension inspired by human cognitive processes through Vision-Aware Text Features. Firstly, we introduce a CLIP Prior module to localize the main object of interest and embed the object heatmap into the query initialization process. Secondly, we propose a combination of two components: Contextual Multimodal Decoder and Meaning Consistency Constraint, to further enhance the coherent and consistent interpretation of language cues with the contextual understanding obtained from the image. Our method achieves significant performance improvements on three benchmark datasets RefCOCO, RefCOCO+ and G-Ref. Project page: \url{https://vatex.hkustvgd.com/}., Comment: This paper is accepted in WACV 2025

When an inverse problem is solved by a gradient-based optimization algorithm, the corresponding forward and adjoint problems, which are introduced to compute the gradient, can be also solved iteratively. The idea of iterating at the same time on the inverse problem unknown and on the forward and adjoint problem solutions yields the concept of one-shot inversion methods. We are especially interested in the case where the inner iterations for the direct and adjoint problems are incomplete, that is, stopped before achieving a high accuracy on their solutions. Here, we focus on general linear inverse problems and generic fixed-point iterations for the associated forward problem. We analyze variants of the so-called multi-step one-shot methods, in particular semi-implicit schemes with a regularization parameter. We establish sufficient conditions on the descent step for convergence, by studying the eigenvalues of the block matrix of the coupled iterations. Several numerical experiments are provided to illustrate the convergence of these methods in comparison with the classical gradient descent, where the forward and adjoint problems are solved exactly by a direct solver instead. We observe that very few inner iterations are enough to guarantee good convergence of the inversion algorithm, even in the presence of noisy data., Comment: arXiv admin note: text overlap with arXiv:2207.10372

This paper studies the wireless scheduling design to coordinate the transmissions of (local) model parameters of federated learning (FL) for a swarm of unmanned aerial vehicles (UAVs). The overall goal of the proposed design is to realize the FL training and aggregation processes with a central aggregator exploiting the sensory data collected by the UAVs but it considers the multi-hop wireless network formed by the UAVs. Such transmissions of model parameters over the UAV-based wireless network potentially cause large transmission delays and overhead. Our proposed framework smartly aggregates local model parameters trained by the UAVs while efficiently transmitting the underlying parameters to the central aggregator in each FL global round. We theoretically show that the proposed scheme achieves minimal delay and communication overhead. Extensive numerical experiments demonstrate the superiority of the proposed scheme compared to other baselines., Comment: accepted to WCNC'24

Large language models (LLMs) have demonstrated a powerful ability to answer various queries as a general-purpose assistant. The continuous multi-modal large language models (MLLM) empower LLMs with the ability to perceive visual signals. The launch of GPT-4 (Generative Pre-trained Transformers) has generated significant interest in the research communities. GPT-4V(ison) has demonstrated significant power in both academia and industry fields, as a focal point in a new artificial intelligence generation. Though significant success was achieved by GPT-4V, exploring MLLMs in domain-specific analysis (e.g., marine analysis) that required domain-specific knowledge and expertise has gained less attention. In this study, we carry out the preliminary and comprehensive case study of utilizing GPT-4V for marine analysis. This report conducts a systematic evaluation of existing GPT-4V, assessing the performance of GPT-4V on marine research and also setting a new standard for future developments in MLLMs. The experimental results of GPT-4V show that the responses generated by GPT-4V are still far away from satisfying the domain-specific requirements of the marine professions. All images and prompts used in this study will be available at https://github.com/hkust-vgd/Marine_GPT-4V_Eval, Comment: 51 pages, 36 figures, Repository: https://github.com/hkust-vgd/Marine_GPT-4V_Eval

Text-to-image diffusion techniques have shown exceptional capability of producing high-quality images from text descriptions. This indicates that there exists a strong correlation between the visual and textual domains. In addition, text-image discriminative models such as CLIP excel in image labelling from text prompts, thanks to the rich and diverse information available from open concepts. In this paper, we leverage these technical advances to solve a challenging problem in computer vision: camouflaged instance segmentation. Specifically, we propose a method built upon a state-of-the-art diffusion model, empowered by open-vocabulary to learn multi-scale textual-visual features for camouflaged object representations. Such cross-domain representations are desirable in segmenting camouflaged objects where visual cues are subtle to distinguish the objects from the background, especially in segmenting novel objects which are not seen in training. We also develop technically supportive components to effectively fuse cross-domain features and engage relevant features towards respective foreground objects. We validate our method and compare it with existing ones on several benchmark datasets of camouflaged instance segmentation and generic open-vocabulary instance segmentation. Experimental results confirm the advances of our method over existing ones. We will publish our code and pre-trained models to support future research., Comment: This work is under review

Assessing the robustness of perception models to covariate shifts and their ability to detect out-of-distribution (OOD) inputs is crucial for safety-critical applications such as autonomous vehicles. By nature of such applications, however, the relevant data is difficult to collect and annotate. In this paper, we show for the first time how synthetic data can be specifically generated to assess comprehensively the real-world reliability of semantic segmentation models. By fine-tuning Stable Diffusion with only in-domain data, we perform zero-shot generation of visual scenes in OOD domains or inpainted with OOD objects. This synthetic data is employed to evaluate the robustness of pretrained segmenters, thereby offering insights into their performance when confronted with real edge cases. Through extensive experiments, we demonstrate a high correlation between the performance of models when evaluated on our synthetic OOD data and when evaluated on real OOD inputs, showing the relevance of such virtual testing. Furthermore, we demonstrate how our approach can be utilized to enhance the calibration and OOD detection capabilities of segmenters. Code and data are made public., Comment: Accepted at ECCV 2024, Project Page: https://valeoai.github.io/publications/genval/ , Code: https://github.com/valeoai/GenVal

Generalization to new domains not seen during training is one of the long-standing challenges in deploying neural networks in real-world applications. Existing generalization techniques either necessitate external images for augmentation, and/or aim at learning invariant representations by imposing various alignment constraints. Large-scale pretraining has recently shown promising generalization capabilities, along with the potential of binding different modalities. For instance, the advent of vision-language models like CLIP has opened the doorway for vision models to exploit the textual modality. In this paper, we introduce a simple framework for generalizing semantic segmentation networks by employing language as the source of randomization. Our recipe comprises three key ingredients: (i) the preservation of the intrinsic CLIP robustness through minimal fine-tuning, (ii) language-driven local style augmentation, and (iii) randomization by locally mixing the source and augmented styles during training. Extensive experiments report state-of-the-art results on various generalization benchmarks. Code is accessible at https://github.com/astra-vision/FAMix ., Comment: CVPR 2024

Data augmentation is a powerful technique to enhance the performance of a deep learning task but has received less attention in 3D deep learning. It is well known that when 3D shapes are sparsely represented with low point density, the performance of the downstream tasks drops significantly. This work explores test-time augmentation (TTA) for 3D point clouds. We are inspired by the recent revolution of learning implicit representation and point cloud upsampling, which can produce high-quality 3D surface reconstruction and proximity-to-surface, respectively. Our idea is to leverage the implicit field reconstruction or point cloud upsampling techniques as a systematic way to augment point cloud data. Mainly, we test both strategies by sampling points from the reconstructed results and using the sampled point cloud as test-time augmented data. We show that both strategies are effective in improving accuracy. We observed that point cloud upsampling for test-time augmentation can lead to more significant performance improvement on downstream tasks such as object classification and segmentation on the ModelNet40, ShapeNet, ScanObjectNN, and SemanticKITTI datasets, especially for sparse point clouds., Comment: This paper is accepted in 3DV 2024

Large language models (LLMs), such as ChatGPT/GPT-4, have proven to be powerful tools in promoting the user experience as an AI assistant. The continuous works are proposing multi-modal large language models (MLLM), empowering LLMs with the ability to sense multiple modality inputs through constructing a joint semantic space (e.g. visual-text space). Though significant success was achieved in LLMs and MLLMs, exploring LLMs and MLLMs in domain-specific applications that required domain-specific knowledge and expertise has been less conducted, especially for \textbf{marine domain}. Different from general-purpose MLLMs, the marine-specific MLLM is required to yield much more \textbf{sensitive}, \textbf{informative}, and \textbf{scientific} responses. In this work, we demonstrate that the existing MLLMs optimized on huge amounts of readily available general-purpose training data show a minimal ability to understand domain-specific intents and then generate informative and satisfactory responses. To address these issues, we propose \textbf{MarineGPT}, the first vision-language model specially designed for the marine domain, unlocking the secrets of the ocean to the public. We present our \textbf{Marine-5M} dataset with more than 5 million marine image-text pairs to inject domain-specific marine knowledge into our model and achieve better marine vision and language alignment. Our MarineGPT not only pushes the boundaries of marine understanding to the general public but also offers a standard protocol for adapting a general-purpose assistant to downstream domain-specific experts. We pave the way for a wide range of marine applications while setting valuable data and pre-trained models for future research in both academic and industrial communities., Comment: work in progress. Code and data will be available at https://github.com/hkust-vgd/MarineGPT

We propose a variational approach for preparing entangled quantum states on quantum computers. The methodology involves training a unitary operation to match with a target unitary using the Fubini-Study distance as a cost function. We employ various gradient-based optimization techniques to enhance performance, including Adam and quantum natural gradient. Our investigation showcases the versatility of different ansatzes featuring a hypergraph structure, enabling the preparation of diverse entanglement target states such as GHZ, W, and absolutely maximally entangled states. Remarkably, the circuit depth scales efficiently with the number of layers and does not depend on the number of qubits. Moreover, we explore the impacts of barren plateaus, readout noise, and error mitigation techniques on the proposed approach. Through our analysis, we demonstrate the effectiveness of the variational algorithm in maximizing the efficiency of quantum state preparation, leveraging low-depth quantum circuits., Comment: 8 pages, 5 figures

Building a video retrieval system that is robust and reliable, especially for the marine environment, is a challenging task due to several factors such as dealing with massive amounts of dense and repetitive data, occlusion, blurriness, low lighting conditions, and abstract queries. To address these challenges, we present MarineVRS, a novel and flexible video retrieval system designed explicitly for the marine domain. MarineVRS integrates state-of-the-art methods for visual and linguistic object representation to enable efficient and accurate search and analysis of vast volumes of underwater video data. In addition, unlike the conventional video retrieval system, which only permits users to index a collection of images or videos and search using a free-form natural language sentence, our retrieval system includes an additional Explainability module that outputs the segmentation masks of the objects that the input query referred to. This feature allows users to identify and isolate specific objects in the video footage, leading to more detailed analysis and understanding of their behavior and movements. Finally, with its adaptability, explainability, accuracy, and scalability, MarineVRS is a powerful tool for marine researchers and scientists to efficiently and accurately process vast amounts of data and gain deeper insights into the behavior and movements of marine species., Comment: Accepted to OCEANS 2023 Limerick. Website: https://marinevrs.hkustvgd.com/

The article deals with anomaly detection of Juniper router logs. Abnormal Juniper router logs include logs that are usually different from the normal operation, and they often reflect the abnormal operation of router devices. To prevent router devices from being damaged and help administrator to grasp the situation of error quickly, detecting abnormal operation soon is very important. In this work, we present a new way to get important features from log data of Juniper router devices and use machine learning method (basing on One-Class SVM model) for anomaly detection. One-Class SVM model requires some knowledge and comprehension about logs of Juniper router devices so that it can analyze, interpret, and test the knowledge ac-quired. We collect log data from a lot of real Juniper router devices and clas-sify them based on our knowledge. Before these logs are used for training and testing the One-Class SVM model, the feature extraction phase for these data was carried out. Finally, with the proposed method, the system errors of the routers were dectected quickly and accurately. This may help our com-pany to reduce the operation cost for the router systems.

Learning models on one labeled dataset that generalize well on another domain is a difficult task, as several shifts might happen between the data domains. This is notably the case for lidar data, for which models can exhibit large performance discrepancies due for instance to different lidar patterns or changes in acquisition conditions. This paper addresses the corresponding Unsupervised Domain Adaptation (UDA) task for semantic segmentation. To mitigate this problem, we introduce an unsupervised auxiliary task of learning an implicit underlying surface representation simultaneously on source and target data. As both domains share the same latent representation, the model is forced to accommodate discrepancies between the two sources of data. This novel strategy differs from classical minimization of statistical divergences or lidar-specific domain adaptation techniques. Our experiments demonstrate that our method achieves a better performance than the current state of the art, both in real-to-real and synthetic-to-real scenarios., Comment: Accepted as spotlight to 3DV 2024. Project repository: github.com/valeoai/SALUDA

Titan, Saturn's largest moon, has a plethora of organic compounds in the atmosphere and on the surface that interact with each other. Cryominerals such as co-crystals may influence the geologic processes and chemical composition of Titan's surface, which in turn informs our understanding of how Titan may have evolved, how the surface is continuing to change, as well as the extent of Titan's habitability. Previous work has shown that a pyridine:acetylene (1:1) co-crystal forms under specific temperatures and experimental conditions; however, this has not yet been demonstrated under Titan-relevant conditions. Our work here demonstrates that the pyridine:acetylene co-crystal is stable from 90 K, Titan's average surface temperature, up to 180 K under an atmosphere of N2. In particular, the co-crystal forms via liquid-solid interactions within minutes upon mixing of the constituents at 150 K, as evidenced by distinct, new Raman bands and band shifts. XRD results indicate moderate anisotropic thermal expansion (about 0.5% - 1.1%) along the three principal axes between 90-150 K. Additionally, the co-crystal is detectable after being exposed to liquid ethane, implying stability in a residual ethane "wetting" scenario on Titan. These results suggest that the pyridine:acetylene co-crystal could form in specific geologic contexts on Titan that allow for warm environments in which liquid pyridine could persist, and as such, this cryomineral may preserve evidence of impact, cryovolcanism, or subsurface transport in surface materials., Comment: 12 pages, 9 figures, 5 tables

Domain adaptation has been vastly investigated in computer vision but still requires access to target images at train time, which might be intractable in some uncommon conditions. In this paper, we propose the task of `Prompt-driven Zero-shot Domain Adaptation', where we adapt a model trained on a source domain using only a general description in natural language of the target domain, i.e., a prompt. First, we leverage a pretrained contrastive vision-language model (CLIP) to optimize affine transformations of source features, steering them towards the target text embedding while preserving their content and semantics. To achieve this, we propose Prompt-driven Instance Normalization (PIN). Second, we show that these prompt-driven augmentations can be used to perform zero-shot domain adaptation for semantic segmentation. Experiments demonstrate that our method significantly outperforms CLIP-based style transfer baselines on several datasets for the downstream task at hand, even surpassing one-shot unsupervised domain adaptation. A similar boost is observed on object detection and image classification. The code is available at https://github.com/astra-vision/PODA ., Comment: Accepted to ICCV 2023, Project Page: https://astra-vision.github.io/PODA/

The 1$^{\text{st}}$ Workshop on Maritime Computer Vision (MaCVi) 2023 focused on maritime computer vision for Unmanned Aerial Vehicles (UAV) and Unmanned Surface Vehicle (USV), and organized several subchallenges in this domain: (i) UAV-based Maritime Object Detection, (ii) UAV-based Maritime Object Tracking, (iii) USV-based Maritime Obstacle Segmentation and (iv) USV-based Maritime Obstacle Detection. The subchallenges were based on the SeaDronesSee and MODS benchmarks. This report summarizes the main findings of the individual subchallenges and introduces a new benchmark, called SeaDronesSee Object Detection v2, which extends the previous benchmark by including more classes and footage. We provide statistical and qualitative analyses, and assess trends in the best-performing methodologies of over 130 submissions. The methods are summarized in the appendix. The datasets, evaluation code and the leaderboard are publicly available at https://seadronessee.cs.uni-tuebingen.de/macvi., Comment: MaCVi 2023 was part of WACV 2023. This report (38 pages) discusses the competition as part of MaCVi