Search

As the key component in multimodal large language models (MLLMs), the ability of the visual encoder greatly affects MLLM's understanding on diverse image content. Although some large-scale pretrained vision encoders such as vision encoders in CLIP and DINOv2 have brought promising performance, we found that there is still no single vision encoder that can dominate various image content understanding, e.g., the CLIP vision encoder leads to outstanding results on general image understanding but poor performance on document or chart content. To alleviate the bias of CLIP vision encoder, we first delve into the inherent behavior of different pre-trained vision encoders and then propose the MoVA, a powerful and novel MLLM, adaptively routing and fusing task-specific vision experts with a coarse-to-fine mechanism. In the coarse-grained stage, we design a context-aware expert routing strategy to dynamically select the most suitable vision experts according to the user instruction, input image, and expertise of vision experts. This benefits from the powerful model function understanding ability of the large language model (LLM) equipped with expert-routing low-rank adaptation (LoRA). In the fine-grained stage, we elaborately conduct the mixture-of-vision-expert adapter (MoV-Adapter) to extract and fuse task-specific knowledge from various experts. This coarse-to-fine paradigm effectively leverages representations from experts based on multimodal context and model expertise, further enhancing the generalization ability. We conduct extensive experiments to evaluate the effectiveness of the proposed approach. Without any bells and whistles, MoVA can achieve significant performance gains over current state-of-the-art methods in a wide range of challenging multimodal benchmarks. Codes and models will be available at https://github.com/TempleX98/MoVA.

Multi-Modal Large Language Models (MLLMs) have demonstrated impressive performance in various VQA tasks. However, they often lack interpretability and struggle with complex visual inputs, especially when the resolution of the input image is high or when the interested region that could provide key information for answering the question is small. To address these challenges, we collect and introduce the large-scale Visual CoT dataset comprising 438k question-answer pairs, annotated with intermediate bounding boxes highlighting key regions essential for answering the questions. Additionally, about 98k pairs of them are annotated with detailed reasoning steps. Importantly, we propose a multi-turn processing pipeline that dynamically focuses on visual inputs and provides interpretable thoughts. We also introduce the related benchmark to evaluate the MLLMs in scenarios requiring specific local region identification. Extensive experiments demonstrate the effectiveness of our framework and shed light on better inference strategies. The Visual CoT dataset, benchmark, and pre-trained models are released to foster further research in this direction., Comment: Code: https://github.com/deepcs233/Visual-CoT

Predicting the future motion of surrounding agents is essential for autonomous vehicles (AVs) to operate safely in dynamic, human-robot-mixed environments. Context information, such as road maps and surrounding agents' states, provides crucial geometric and semantic information for motion behavior prediction. To this end, recent works explore two-stage prediction frameworks where coarse trajectories are first proposed, and then used to select critical context information for trajectory refinement. However, they either incur a large amount of computation or bring limited improvement, if not both. In this paper, we introduce a novel scenario-adaptive refinement strategy, named SmartRefine, to refine prediction with minimal additional computation. Specifically, SmartRefine can comprehensively adapt refinement configurations based on each scenario's properties, and smartly chooses the number of refinement iterations by introducing a quality score to measure the prediction quality and remaining refinement potential of each scenario. SmartRefine is designed as a generic and flexible approach that can be seamlessly integrated into most state-of-the-art motion prediction models. Experiments on Argoverse (1 & 2) show that our method consistently improves the prediction accuracy of multiple state-of-the-art prediction models. Specifically, by adding SmartRefine to QCNet, we outperform all published ensemble-free works on the Argoverse 2 leaderboard (single agent track) at submission. Comprehensive studies are also conducted to ablate design choices and explore the mechanism behind multi-iteration refinement. Codes are available at https://github.com/opendilab/SmartRefine/, Comment: Camera-ready version for CVPR 2024

We propose SPHINX-X, an extensive Multimodality Large Language Model (MLLM) series developed upon SPHINX. To improve the architecture and training efficiency, we modify the SPHINX framework by removing redundant visual encoders, bypassing fully-padded sub-images with skip tokens, and simplifying multi-stage training into a one-stage all-in-one paradigm. To fully unleash the potential of MLLMs, we assemble a comprehensive multi-domain and multimodal dataset covering publicly available resources in language, vision, and vision-language tasks. We further enrich this collection with our curated OCR intensive and Set-of-Mark datasets, extending the diversity and generality. By training over different base LLMs including TinyLlama1.1B, InternLM2-7B, LLaMA2-13B, and Mixtral8x7B, we obtain a spectrum of MLLMs that vary in parameter size and multilingual capabilities. Comprehensive benchmarking reveals a strong correlation between the multi-modal performance with the data and parameter scales. Code and models are released at https://github.com/Alpha-VLLM/LLaMA2-Accessory, Comment: Accepted by ICML 2024. Code and models are released at https://github.com/Alpha-VLLM/LLaMA2-Accessory

Efficiently capturing multi-scale information and building long-range dependencies among pixels are essential for medical image segmentation because of the various sizes and shapes of the lesion regions or organs. In this paper, we present Multi-scale Cross-axis Attention (MCA) to solve the above challenging issues based on the efficient axial attention. Instead of simply connecting axial attention along the horizontal and vertical directions sequentially, we propose to calculate dual cross attentions between two parallel axial attentions to capture global information better. To process the significant variations of lesion regions or organs in individual sizes and shapes, we also use multiple convolutions of strip-shape kernels with different kernel sizes in each axial attention path to improve the efficiency of the proposed MCA in encoding spatial information. We build the proposed MCA upon the MSCAN backbone, yielding our network, termed MCANet. Our MCANet with only 4M+ parameters performs even better than most previous works with heavy backbones (e.g., Swin Transformer) on four challenging tasks, including skin lesion segmentation, nuclei segmentation, abdominal multi-organ segmentation, and polyp segmentation. Code is available at https://github.com/haoshao-nku/medical_seg.

We present a new boundary sensitive framework for polyp segmentation, called Polyper. Our method is motivated by a clinical approach that seasoned medical practitioners often leverage the inherent features of interior polyp regions to tackle blurred boundaries.Inspired by this, we propose explicitly leveraging polyp regions to bolster the model's boundary discrimination capability while minimizing computation. Our approach first extracts boundary and polyp regions from the initial segmentation map through morphological operators. Then, we design the boundary sensitive attention that concentrates on augmenting the features near the boundary regions using the interior polyp regions's characteristics to generate good segmentation results. Our proposed method can be seamlessly integrated with classical encoder networks, like ResNet-50, MiT-B1, and Swin Transformer. To evaluate the effectiveness of Polyper, we conduct experiments on five publicly available challenging datasets, and receive state-of-the-art performance on all of them. Code is available at https://github.com/haoshao-nku/medical_seg.git., Comment: Accepted to AAAI 2024

Despite significant recent progress in the field of autonomous driving, modern methods still struggle and can incur serious accidents when encountering long-tail unforeseen events and challenging urban scenarios. On the one hand, large language models (LLM) have shown impressive reasoning capabilities that approach "Artificial General Intelligence". On the other hand, previous autonomous driving methods tend to rely on limited-format inputs (e.g. sensor data and navigation waypoints), restricting the vehicle's ability to understand language information and interact with humans. To this end, this paper introduces LMDrive, a novel language-guided, end-to-end, closed-loop autonomous driving framework. LMDrive uniquely processes and integrates multi-modal sensor data with natural language instructions, enabling interaction with humans and navigation software in realistic instructional settings. To facilitate further research in language-based closed-loop autonomous driving, we also publicly release the corresponding dataset which includes approximately 64K instruction-following data clips, and the LangAuto benchmark that tests the system's ability to handle complex instructions and challenging driving scenarios. Extensive closed-loop experiments are conducted to demonstrate LMDrive's effectiveness. To the best of our knowledge, we're the very first work to leverage LLMs for closed-loop end-to-end autonomous driving. Codes, models, and datasets can be found at https://github.com/opendilab/LMDrive, Comment: project page: https://hao-shao.com/projects/lmdrive.html

The large-scale deployment of autonomous vehicles is yet to come, and one of the major remaining challenges lies in urban dense traffic scenarios. In such cases, it remains challenging to predict the future evolution of the scene and future behaviors of objects, and to deal with rare adverse events such as the sudden appearance of occluded objects. In this paper, we present ReasonNet, a novel end-to-end driving framework that extensively exploits both temporal and global information of the driving scene. By reasoning on the temporal behavior of objects, our method can effectively process the interactions and relationships among features in different frames. Reasoning about the global information of the scene can also improve overall perception performance and benefit the detection of adverse events, especially the anticipation of potential danger from occluded objects. For comprehensive evaluation on occlusion events, we also release publicly a driving simulation benchmark DriveOcclusionSim consisting of diverse occlusion events. We conduct extensive experiments on multiple CARLA benchmarks, where our model outperforms all prior methods, ranking first on the sensor track of the public CARLA Leaderboard., Comment: CVPR 2023

When autonomous vehicles are deployed on public roads, they will encounter countless and diverse driving situations. Many manually designed driving policies are difficult to scale to the real world. Fortunately, reinforcement learning has shown great success in many tasks by automatic trial and error. However, when it comes to autonomous driving in interactive dense traffic, RL agents either fail to learn reasonable performance or necessitate a large amount of data. Our insight is that when humans learn to drive, they will 1) make decisions over the high-level skill space instead of the low-level control space and 2) leverage expert prior knowledge rather than learning from scratch. Inspired by this, we propose ASAP-RL, an efficient reinforcement learning algorithm for autonomous driving that simultaneously leverages motion skills and expert priors. We first parameterized motion skills, which are diverse enough to cover various complex driving scenarios and situations. A skill parameter inverse recovery method is proposed to convert expert demonstrations from control space to skill space. A simple but effective double initialization technique is proposed to leverage expert priors while bypassing the issue of expert suboptimality and early performance degradation. We validate our proposed method on interactive dense-traffic driving tasks given simple and sparse rewards. Experimental results show that our method can lead to higher learning efficiency and better driving performance relative to previous methods that exploit skills and priors differently. Code is open-sourced to facilitate further research., Comment: Robotics: Science and Systems (RSS 2023)

We investigate the effect of a lower index oxide layer inclusion within a highly doped silica glass slot waveguide for optimized supercontinuum generation at telecom wavelengths. By controlling the thickness of the oxide slot, we demonstrate that one can engineer the waveguide dispersion profile so that to obtain supercontinua with vastly different spectral broadening dynamics and bandwidths. Using this approach, we designed and fabricated a waveguide with a low and flat dispersion profile of less than 43 ps/km/nm across a wavelength range spanning over 1000 nm. We show that, when pumped at the telecom C-band, we can generate a supercontinuum that spans over 1.5 octave, from approximately 817 nm to 2183 nm. The numerical simulations, whose parameters are derived from the measured waveguide dimension and material indices, exhibit good agreement with experimental measurements, where one can observe both a qualitative and quantitative match in the supercontinuum overall spectrum and specific features (e.g. soliton and dispersive wave locations). This study represents an important step forward in the control and manipulation of dispersive and nonlinear dynamics in highly doped silica glass waveguides, paving the way towards advanced on-chip broadband light manipulation.

Large-scale deployment of autonomous vehicles has been continually delayed due to safety concerns. On the one hand, comprehensive scene understanding is indispensable, a lack of which would result in vulnerability to rare but complex traffic situations, such as the sudden emergence of unknown objects. However, reasoning from a global context requires access to sensors of multiple types and adequate fusion of multi-modal sensor signals, which is difficult to achieve. On the other hand, the lack of interpretability in learning models also hampers the safety with unverifiable failure causes. In this paper, we propose a safety-enhanced autonomous driving framework, named Interpretable Sensor Fusion Transformer(InterFuser), to fully process and fuse information from multi-modal multi-view sensors for achieving comprehensive scene understanding and adversarial event detection. Besides, intermediate interpretable features are generated from our framework, which provide more semantics and are exploited to better constrain actions to be within the safe sets. We conducted extensive experiments on CARLA benchmarks, where our model outperforms prior methods, ranking the first on the public CARLA Leaderboard. Our code will be made available at https://github.com/opendilab/InterFuser, Comment: Accepted at CoRL 2022

Learning-based pre-simulation (i.e., layout-to-fabrication) models have been proposed to predict the fabrication-induced shape deformation from an IC layout to its fabricated circuit. Such models are usually driven by pairwise learning, involving a training set of layout patterns and their reference shape images after fabrication. However, it is expensive and time-consuming to collect the reference shape images of all layout clips for model training and updating. To address the problem, we propose a deep learning-based layout novelty detection scheme to identify novel (unseen) layout patterns, which cannot be well predicted by a pre-trained pre-simulation model. We devise a global-local novelty scoring mechanism to assess the potential novelty of a layout by exploiting two subnetworks: an autoencoder and a pretrained pre-simulation model. The former characterizes the global structural dissimilarity between a given layout and training samples, whereas the latter extracts a latent code representing the fabrication-induced local deformation. By integrating the global dissimilarity with the local deformation boosted by a self-attention mechanism, our model can accurately detect novelties without the ground-truth circuit shapes of test samples. Based on the detected novelties, we further propose two active-learning strategies to sample a reduced amount of representative layouts most worthy to be fabricated for acquiring their ground-truth circuit shapes. Experimental results demonstrate i) our method's effectiveness in layout novelty detection, and ii) our active-learning strategies' ability in selecting representative novel layouts for keeping a learning-based pre-simulation model updated.

The transformer architectures, based on self-attention mechanism and convolution-free design, recently found superior performance and booming applications in computer vision. However, the discontinuous patch-wise tokenization process implicitly introduces jagged artifacts into attention maps, arising the traditional problem of aliasing for vision transformers. Aliasing effect occurs when discrete patterns are used to produce high frequency or continuous information, resulting in the indistinguishable distortions. Recent researches have found that modern convolution networks still suffer from this phenomenon. In this work, we analyze the uncharted problem of aliasing in vision transformer and explore to incorporate anti-aliasing properties. Specifically, we propose a plug-and-play Aliasing-Reduction Module(ARM) to alleviate the aforementioned issue. We investigate the effectiveness and generalization of the proposed method across multiple tasks and various vision transformer families. This lightweight design consistently attains a clear boost over several famous structures. Furthermore, our module also improves data efficiency and robustness of vision transformers., Comment: Accepted to NeurIPS 2021

This paper investigates the cointegration between possible determinants of crude oil futures prices during the COVID-19 pandemic period. We perform comparative analysis of WTI and newly-launched Shanghai crude oil futures (SC) via the Autoregressive Distributed Lag (ARDL) model and Quantile Autoregressive Distributed Lag (QARDL) model. The empirical results confirm that economic policy uncertainty, stock markets, interest rates and coronavirus panic are important drivers of WTI futures prices. Our findings also suggest that the US and China's stock markets play vital roles in movements of SC futures prices. Meanwhile, CSI300 stock index has a significant positive short-run impact on SC futures prices while S\&P500 prices possess a positive nexus with SC futures prices both in long-run and short-run. Overall, these empirical evidences provide practical implications for investors and policymakers.

This paper pays close attention to the cross-modality visible-infrared person re-identification (VI Re-ID) task, which aims to match pedestrian samples between visible and infrared modes. In order to reduce the modality-discrepancy between samples from different cameras, most existing works usually use constraints based on Euclidean metric. Because of the Euclidean based distance metric strategy cannot effectively measure the internal angles between the embedded vectors, the existing solutions cannot learn the angularly discriminative feature embedding. Since the most important factor affecting the classification task based on embedding vector is whether there is an angularly discriminative feature space, in this paper, we present a new loss function called Enumerate Angular Triplet (EAT) loss. Also, motivated by the knowledge distillation, to narrow down the features between different modalities before feature embedding, we further present a novel Cross-Modality Knowledge Distillation (CMKD) loss. Benefit from the above two considerations, the embedded features are discriminative enough in a way to tackle modality-discrepancy problem. The experimental results on RegDB and SYSU-MM01 datasets have demonstrated that the proposed method is superior to the other most advanced methods in terms of impressive performance. Code is available at https://github.com/IVIPLab/LCCRF., Comment: World Wide Web Journal, 20 pages, 6 figures

Label noise in training data can significantly degrade a model's generalization performance for supervised learning tasks. Here we focus on the problem that noisy labels are primarily mislabeled samples, which tend to be concentrated near decision boundaries, rather than uniformly distributed, and whose features should be equivocal. To address the problem, we propose an ensemble learning method to correct noisy labels by exploiting the local structures of feature manifolds. Different from typical ensemble strategies that increase the prediction diversity among sub-models via certain loss terms, our method trains sub-models on disjoint subsets, each being a union of the nearest-neighbors of randomly selected seed samples on the data manifold. As a result, each sub-model can learn a coarse representation of the data manifold along with a corresponding graph. Moreover, only a limited number of sub-models will be affected by locally-concentrated noisy labels. The constructed graphs are used to suggest a series of label correction candidates, and accordingly, our method derives label correction results by voting down inconsistent suggestions. Our experiments on real-world noisy label datasets demonstrate the superiority of the proposed method over existing state-of-the-arts.

Recently, falsified images have been found in papers involved in research misconducts. However, although there have been many image forgery detection methods, none of them was designed for molecular-biological experiment images. In this paper, we proposed a fast blind inquiry method, named FBI$_{GEL}$, for integrity of images obtained from two common sorts of molecular experiments, i.e., western blot (WB) and polymerase chain reaction (PCR). Based on an optimized pseudo-background capable of highlighting local residues, FBI$_{GEL}$ can reveal traceable vestiges suggesting inappropriate local modifications on WB/PCR images. Additionally, because the optimized pseudo-background is derived according to a closed-form solution, FBI$_{GEL}$ is computationally efficient and thus suitable for large scale inquiry tasks for WB/PCR image integrity. We applied FBI$_{GEL}$ on several papers questioned by the public on \textbf{PUBPEER}, and our results show that figures of those papers indeed contain doubtful unnatural patterns., Comment: This version is an extension of Prof. Shao's previous conference paper (IEEE GlobalSIP 2018): "Unveiling Vestiges of Man-Made Modifications on Molecular-Biological Experiment Images." (https://doi.org/10.1109/GlobalSIP.2018.8646594)

This technical report presents an overview of our solution used in the submission to ActivityNet Challenge 2020 Task 1 (\textbf{temporal action localization/detection}). Temporal action localization requires to not only precisely locate the temporal boundaries of action instances, but also accurately classify the untrimmed videos into specific categories. In this paper, we decouple the temporal action localization task into two stages (i.e. proposal generation and classification) and enrich the proposal diversity through exhaustively exploring the influences of multiple components from different but complementary perspectives. Specifically, in order to generate high-quality proposals, we consider several factors including the video feature encoder, the proposal generator, the proposal-proposal relations, the scale imbalance, and ensemble strategy. Finally, in order to obtain accurate detections, we need to further train an optimal video classifier to recognize the generated proposals. Our proposed scheme achieves the state-of-the-art performance on the temporal action localization task with \textbf{42.26} average mAP on the challenge testing set., Comment: Submitted to CVPR workshop of ActivityNet Challenge 2020

This technical report introduces our winning solution to the spatio-temporal action localization track, AVA-Kinetics Crossover, in ActivityNet Challenge 2020. Our entry is mainly based on Actor-Context-Actor Relation Network. We describe technical details for the new AVA-Kinetics dataset, together with some experimental results. Without any bells and whistles, we achieved 39.62 mAP on the test set of AVA-Kinetics, which outperforms other entries by a large margin. Code will be available at: https://github.com/Siyu-C/ACAR-Net., Comment: arXiv admin note: substantial text overlap with arXiv:2006.07976

In this technical report, we briefly introduce the solutions of our team 'Efficient' for the Multi-Moments in Time challenge in ICCV 2019. We first conduct several experiments with popular Image-Based action recognition methods TRN, TSN, and TSM. Then a novel temporal interlacing network is proposed towards fast and accurate recognition. Besides, the SlowFast network and its variants are explored. Finally, we ensemble all the above models and achieve 67.22\% on the validation set and 60.77\% on the test set, which ranks 1st on the final leaderboard. In addition, we release a new code repository for video understanding which unifies state-of-the-art 2D and 3D methods based on PyTorch. The solution of the challenge is also included in the repository, which is available at https://github.com/Sense-X/X-Temporal.

Knowledge graph models world knowledge as concepts, entities, and the relationships between them, which has been widely used in many real-world tasks. CCKS 2019 held an evaluation track with 6 tasks and attracted more than 1,600 teams. In this paper, we give an overview of the knowledge graph evaluation tract at CCKS 2019. By reviewing the task definition, successful methods, useful resources, good strategies and research challenges associated with each task in CCKS 2019, this paper can provide a helpful reference for developing knowledge graph applications and conducting future knowledge graph researches., Comment: 21 pages, in Chinese, 9 figures and 17 tables, CCKS 2019 held an evaluation track about knowledge graph with 6 tasks and attracted more than 1,600 teams

The performance of a convolutional neural network (CNN) based face recognition model largely relies on the richness of labelled training data. Collecting a training set with large variations of a face identity under different poses and illumination changes, however, is very expensive, making the diversity of within-class face images a critical issue in practice. In this paper, we propose a 3D model-assisted domain-transferred face augmentation network (DotFAN) that can generate a series of variants of an input face based on the knowledge distilled from existing rich face datasets collected from other domains. DotFAN is structurally a conditional CycleGAN but has two additional subnetworks, namely face expert network (FEM) and face shape regressor (FSR), for latent code control. While FSR aims to extract face attributes, FEM is designed to capture a face identity. With their aid, DotFAN can learn a disentangled face representation and effectively generate face images of various facial attributes while preserving the identity of augmented faces. Experiments show that DotFAN is beneficial for augmenting small face datasets to improve their within-class diversity so that a better face recognition model can be learned from the augmented dataset., Comment: 12 pages, 10 figures

We propose a deep learning-based data-driven framework consisting of two convolutional neural networks: i) LithoNet that predicts the shape deformations on a circuit due to IC fabrication, and ii) OPCNet that suggests IC layout corrections to compensate for such shape deformations. By learning the shape correspondences between pairs of layout design patterns and their scanning electron microscope (SEM) images of the product wafer thereof, given an IC layout pattern, LithoNet can mimic the fabrication process to predict its fabricated circuit shape. Furthermore, LithoNet can take the wafer fabrication parameters as a latent vector to model the parametric product variations that can be inspected on SEM images. Besides, traditional optical proximity correction (OPC) methods used to suggest a correction on a lithographic photomask is computationally expensive. Our proposed OPCNet mimics the OPC procedure and efficiently generates a corrected photomask by collaborating with LithoNet to examine if the shape of a fabricated circuit optimally matches its original layout design. As a result, the proposed LithoNet-OPCNet framework can not only predict the shape of a fabricated IC from its layout pattern, but also suggests a layout correction according to the consistency between the predicted shape and the given layout. Experimental results with several benchmark layout patterns demonstrate the effectiveness of the proposed method., Comment: 14 pages, 16 figures

For a long time, the vision community tries to learn the spatio-temporal representation by combining convolutional neural network together with various temporal models, such as the families of Markov chain, optical flow, RNN and temporal convolution. However, these pipelines consume enormous computing resources due to the alternately learning process for spatial and temporal information. One natural question is whether we can embed the temporal information into the spatial one so the information in the two domains can be jointly learned once-only. In this work, we answer this question by presenting a simple yet powerful operator -- temporal interlacing network (TIN). Instead of learning the temporal features, TIN fuses the two kinds of information by interlacing spatial representations from the past to the future, and vice versa. A differentiable interlacing target can be learned to control the interlacing process. In this way, a heavy temporal model is replaced by a simple interlacing operator. We theoretically prove that with a learnable interlacing target, TIN performs equivalently to the regularized temporal convolution network (r-TCN), but gains 4% more accuracy with 6x less latency on 6 challenging benchmarks. These results push the state-of-the-art performances of video understanding by a considerable margin. Not surprising, the ensemble model of the proposed TIN won the $1^{st}$ place in the ICCV19 - Multi Moments in Time challenge. Code is made available to facilitate further research at https://github.com/deepcs233/TIN, Comment: Accepted to AAAI 2020. Winning entry of ICCV Multi-Moments in Time Challenge 2019. Code is available at https://github.com/deepcs233/TIN