Search

Large multimodal models (LMMs) excel in adhering to human instructions. However, self-contradictory instructions may arise due to the increasing trend of multimodal interaction and context length, which is challenging for language beginners and vulnerable populations. We introduce the Self-Contradictory Instructions benchmark to evaluate the capability of LMMs in recognizing conflicting commands. It comprises 20,000 conflicts, evenly distributed between language and vision paradigms. It is constructed by a novel automatic dataset creation framework, which expedites the process and enables us to encompass a wide range of instruction forms. Our comprehensive evaluation reveals current LMMs consistently struggle to identify multimodal instruction discordance due to a lack of self-awareness. Hence, we propose the Cognitive Awakening Prompting to inject cognition from external, largely enhancing dissonance detection. The dataset and code are here: https://selfcontradiction.github.io/., Comment: Accepted by the 18th European Conference on Computer Vision ECCV 2024

We introduce the H\"older regularity condition for harmonic functions on metric measure spaces and prove that under mild volume regular condition and upper heat kernel estimate, the H\"older regularity condition, the weak Bakry-\'Emery non-negative curvature condition, the heat kernel H\"older continuity with or without exponential terms and the heat kernel near-diagonal lower bound are equivalent. As applications, firstly, we prove the validity of the so-called generalized reverse H\"older inequality on the Sierpi\'nski carpet cable system, which was left open by Devyver, Russ, Yang (Int. Math. Res. Not. IMRN (2023), no. 18, 15537-15583). Secondly, we prove that two-sided heat kernel estimates alone imply gradient estimate for the heat kernel on strongly recurrent fractal-like cable systems, which improves the main results of the aforementioned paper. Thirdly, we obtain H\"older (Lipschitz) estimate for heat kernel on general metric measure spaces, which extends the classical Li-Yau gradient estimate for heat kernel on Riemannian manifolds., Comment: 36 pages, some remarks and references added

Escape of active agents from metastable states is of great interest in statistical and biological physics. In this study, we investigate the escape of a flexible active ring, composed of active Brownian particles, from a flat attractive surface using Brownian dynamics simulations. To systematically explore the effects of activity, persistence time, and the shape of attractive potentials, we calculate escape time and effective temperature. We observe two distinct escape mechanisms: Kramers-like thermal activation at small persistence times and the maximal force problem at large persistence time, where escape time is determined by persistence time. The escape time explicitly depends on the shape of the potential barrier at high activity and large persistence time. Moreover, when the propulsion force is biased along the ring's contour, escape becomes more difficult and is primarily driven by thermal noise. Our findings highlight that, despite its intricate configuration, the active ring can be effectively modeled as a self-propelled Brownian particle when studying its escape from a smooth surface.

It has been supposed that the interplay of elasticity and activity plays a key role in triggering the non-equilibrium behaviors in biological systems. However, the experimental model system is missing to investigate the spatiotemporally dynamical phenomena. Here, a model system of an active chain, where active eccentric-disks are linked by a spring, is designed to study the interplay of activity, elasticity, and friction. Individual active chain exhibits longitudinal and transverse motion, however, it starts to self-rotate when pinning one end, and self-beats when clamping one end. Additionally, our eccentric-disk model can qualitatively reproduce such behaviors and explain the unusual self-rotation of the first disk around its geometric center. Further, the structure and dynamics of long chains were studied via simulations without steric interactions. It was found that hairpin conformation emerges in free motion, while in the constrained motions, the rotational and beating frequencies scale with the flexure number (the ratio of self-propelling force to bending rigidity), ~4/3. Scaling analysis suggests that it results from the balance between activity and energy dissipation. Our findings show that topological constraints play a vital role in non-equilibrium synergy behavior.

Joint Detection and Embedding (JDE) trackers have demonstrated excellent performance in Multi-Object Tracking (MOT) tasks by incorporating the extraction of appearance features as auxiliary tasks through embedding Re-Identification task (ReID) into the detector, achieving a balance between inference speed and tracking performance. However, solving the competition between the detector and the feature extractor has always been a challenge. Meanwhile, the issue of directly embedding the ReID task into MOT has remained unresolved. The lack of high discriminability in appearance features results in their limited utility. In this paper, a new learning approach using cross-correlation to capture temporal information of objects is proposed. The feature extraction network is no longer trained solely on appearance features from each frame but learns richer motion features by utilizing feature heatmaps from consecutive frames, which addresses the challenge of inter-class feature similarity. Furthermore, our learning approach is applied to a more lightweight feature extraction network, and treat the feature matching scores as strong cues rather than auxiliary cues, with an appropriate weight calculation to reflect the compatibility between our obtained features and the MOT task. Our tracker, named TCBTrack, achieves state-of-the-art performance on multiple public benchmarks, i.e., MOT17, MOT20, and DanceTrack datasets. Specifically, on the DanceTrack test set, we achieve 56.8 HOTA, 58.1 IDF1 and 92.5 MOTA, making it the best online tracker capable of achieving real-time performance. Comparative evaluations with other trackers prove that our tracker achieves the best balance between speed, robustness and accuracy. Code is available at https://github.com/yfzhang1214/TCBTrack., Comment: A submission to IJCV

Recent advances in 4D generation mainly focus on generating 4D content by distilling pre-trained text or single-view image-conditioned models. It is inconvenient for them to take advantage of various off-the-shelf 3D assets with multi-view attributes, and their results suffer from spatiotemporal inconsistency owing to the inherent ambiguity in the supervision signals. In this work, we present Animate3D, a novel framework for animating any static 3D model. The core idea is two-fold: 1) We propose a novel multi-view video diffusion model (MV-VDM) conditioned on multi-view renderings of the static 3D object, which is trained on our presented large-scale multi-view video dataset (MV-Video). 2) Based on MV-VDM, we introduce a framework combining reconstruction and 4D Score Distillation Sampling (4D-SDS) to leverage the multi-view video diffusion priors for animating 3D objects. Specifically, for MV-VDM, we design a new spatiotemporal attention module to enhance spatial and temporal consistency by integrating 3D and video diffusion models. Additionally, we leverage the static 3D model's multi-view renderings as conditions to preserve its identity. For animating 3D models, an effective two-stage pipeline is proposed: we first reconstruct motions directly from generated multi-view videos, followed by the introduced 4D-SDS to refine both appearance and motion. Benefiting from accurate motion learning, we could achieve straightforward mesh animation. Qualitative and quantitative experiments demonstrate that Animate3D significantly outperforms previous approaches. Data, code, and models will be open-released., Comment: Project Page: https://animate3d.github.io/

Efficient and robust grasp pose detection is vital for robotic manipulation. For general 6 DoF grasping, conventional methods treat all points in a scene equally and usually adopt uniform sampling to select grasp candidates. However, we discover that ignoring where to grasp greatly harms the speed and accuracy of current grasp pose detection methods. In this paper, we propose "graspness", a quality based on geometry cues that distinguishes graspable areas in cluttered scenes. A look-ahead searching method is proposed for measuring the graspness and statistical results justify the rationality of our method. To quickly detect graspness in practice, we develop a neural network named cascaded graspness model to approximate the searching process. Extensive experiments verify the stability, generality and effectiveness of our graspness model, allowing it to be used as a plug-and-play module for different methods. A large improvement in accuracy is witnessed for various previous methods after equipping our graspness model. Moreover, we develop GSNet, an end-to-end network that incorporates our graspness model for early filtering of low-quality predictions. Experiments on a large-scale benchmark, GraspNet-1Billion, show that our method outperforms previous arts by a large margin (30+ AP) and achieves a high inference speed. The library of GSNet has been integrated into AnyGrasp, which is at https://github.com/graspnet/anygrasp_sdk., Comment: ICCV 2021

Recently, the Hofstadter spectrum of a twisted $\mathrm{WSe_2/MoSe_2}$ heterobilayer has been observed in experiment [C. R. Kometter, et al. Nat.Phys.19, 1861 (2023)], but the origin of Hofstadter states remains unclear. Here, we present a comprehensive theoretical interpretation of the observed Hofstadter states by calculating its accurate Hofstadter spectrum. We point out that the valley Zeeman effect, a unique feature of the transition metal dichalcogenide (TMD) materials, plays a crucial role in determining the shape of the Hofstadter spectrum, due to the narrow bandwidth of the moir\'e bands. This is distinct from the graphene-based moir\'e systems. We further predict that the Hofstadter spectrum of the moir\'e flat band, which was not observed in experiment, can be observed in the same system with a larger twist angle $2^\circ\lesssim\theta \lesssim 3^\circ$. Our theory paves the way for further studies of the interplay between the Hofstadter states and correlated insulting states in such moir\'e lattice systems., Comment: 7 pages, 4 figures

In this paper, we establish the existence of $p$-energy norms and the corresponding $p$-energy measures for scale-irregular Vicsek sets, which may lack self-similarity. We also investigate the characterizations of $p$-energy norms in terms of Besov-Lipschitz norms, with their weak monotonicity and the corresponding Bourgain-Brezis-Mironescu convergence., Comment: 34 pages, 2 figures

The long-awaited fractional quantum anomalous Hall (FQAH) effect recently has been observed in the twisted $\mathrm{MoTe_2}$ homobilayers, causing a great sensation. Here, we theoretically investigate the moir\'e band structures of a closely related system, the alternating twisted multilayer $\mathrm{MoTe_2}$ (ATML-$\mathrm{MoTe_2}$), where the adjacent layers have opposite twist angles. We illustrate that such ATML-$\mathrm{MoTe_2}$ is a very unique moir\'e system, exhibiting multiple topological flat bands highly controllable by the layer number and twist angle, which is not only an ideal platform to simulate Hubbard model, but also may host FQAH states. Specifically, an N-layer ATML-$\mathrm{MoTe_2}$ ($N \geq 3$) always possesses $N-2$ topological flat bands near Fermi energy $E_f$, which has an odd-even dependent decomposition rule to understand the behaviors of the moir\'e flat bands. We predict three intriguing examples: (1) The AT3L-$\mathrm{MoTe_2}$ ($N=3$) has one isolated moir\'e flat band, which corresponds to a triangular lattice Hubbard model, resembling the twisted TMD heterobilayers. (2) The AT4L-$\mathrm{MoTe_2}$ ($N=4$) has two topological flat bands that are very similar to the twisted $\mathrm{MoTe_2}$ homobilayers, implying the possible existence of FQAH states. (3) When $N>4$, the giant density of states (DOS) induced by the multiple moir\'e flat bands may induce exotic correlated states., Comment: 11 pages, 5 figures

Masked image modeling (MIM) pre-training for large-scale vision transformers (ViTs) has enabled promising downstream performance on top of the learned self-supervised ViT features. In this paper, we question if the \textit{extremely simple} lightweight ViTs' fine-tuning performance can also benefit from this pre-training paradigm, which is considerably less studied yet in contrast to the well-established lightweight architecture design methodology. We use an observation-analysis-solution flow for our study. We first systematically observe different behaviors among the evaluated pre-training methods with respect to the downstream fine-tuning data scales. Furthermore, we analyze the layer representation similarities and attention maps across the obtained models, which clearly show the inferior learning of MIM pre-training on higher layers, leading to unsatisfactory transfer performance on data-insufficient downstream tasks. This finding is naturally a guide to designing our distillation strategies during pre-training to solve the above deterioration problem. Extensive experiments have demonstrated the effectiveness of our approach. Our pre-training with distillation on pure lightweight ViTs with vanilla/hierarchical design ($5.7M$/$6.5M$) can achieve $79.4\%$/$78.9\%$ top-1 accuracy on ImageNet-1K. It also enables SOTA performance on the ADE20K segmentation task ($42.8\%$ mIoU) and LaSOT tracking task ($66.1\%$ AUC) in the lightweight regime. The latter even surpasses all the current SOTA lightweight CPU-realtime trackers., Comment: A submission to IJCV

In this paper, we propose a new image-based visual place recognition (VPR) framework by exploiting the structural cues in bird's-eye view (BEV) from a single monocular camera. The motivation arises from two key observations about place recognition methods based on both appearance and structure: 1) For the methods relying on LiDAR sensors, the integration of LiDAR in robotic systems has led to increased expenses, while the alignment of data between different sensors is also a major challenge. 2) Other image-/camera-based methods, involving integrating RGB images and their derived variants (eg, pseudo depth images, pseudo 3D point clouds), exhibit several limitations, such as the failure to effectively exploit the explicit spatial relationships between different objects. To tackle the above issues, we design a new BEV-enhanced VPR framework, namely BEV$^2$PR, generating a composite descriptor with both visual cues and spatial awareness based on a single camera. The key points lie in: 1) We use BEV features as an explicit source of structural knowledge in constructing global features. 2) The lower layers of the pre-trained backbone from BEV generation are shared for visual and structural streams in VPR, facilitating the learning of fine-grained local features in the visual stream. 3) The complementary visual and structural features can jointly enhance VPR performance. Our BEV$^2$PR framework enables consistent performance improvements over several popular aggregation modules for RGB global features. The experiments on our collected VPR-NuScenes dataset demonstrate an absolute gain of 2.47% on Recall@1 for the strong Conv-AP baseline to achieve the best performance in our setting, and notably, a 18.06% gain on the hard set. The code and dataset will be available at https://github.com/FudongGe/BEV2PR., Comment: Accepted at IROS 2024 as Oral Presentation. Code available at https://github.com/FudongGe/BEV2PR

In this study, we develop a multiple-generative agent system to simulate community decision-making for the redevelopment of Kendall Square's Volpe building. Drawing on interviews with local stakeholders, our simulations incorporated varying degrees of communication, demographic data, and life values in the agent prompts. The results revealed that communication among agents improved collective reasoning, while the inclusion of demographic and life values led to more distinct opinions. These findings highlight the potential application of AI in understanding complex social interactions and decision-making processes, offering valuable insights for urban planning and community engagement in diverse settings like Kendall Square.

The advent of foundation models (FMs) as an emerging suite of AI techniques has struck a wave of opportunities in computational healthcare. The interactive nature of these models, guided by pre-training data and human instructions, has ignited a data-centric AI paradigm that emphasizes better data characterization, quality, and scale. In healthcare AI, obtaining and processing high-quality clinical data records has been a longstanding challenge, ranging from data quantity, annotation, patient privacy, and ethics. In this survey, we investigate a wide range of data-centric approaches in the FM era (from model pre-training to inference) towards improving the healthcare workflow. We discuss key perspectives in AI security, assessment, and alignment with human values. Finally, we offer a promising outlook of FM-based analytics to enhance the performance of patient outcome and clinical workflow in the evolving landscape of healthcare and medicine. We provide an up-to-date list of healthcare-related foundation models and datasets at https://github.com/Yunkun-Zhang/Data-Centric-FM-Healthcare .

Point cloud-based 3D object tracking is an important task in autonomous driving. Though great advances regarding Siamese-based 3D tracking have been made recently, it remains challenging to learn the correlation between the template and search branches effectively with the sparse LIDAR point cloud data. Instead of performing correlation of the two branches at just one point in the network, in this paper, we present a multi-correlation Siamese Transformer network that has multiple stages and carries out feature correlation at the end of each stage based on sparse pillars. More specifically, in each stage, self-attention is first applied to each branch separately to capture the non-local context information. Then, cross-attention is used to inject the template information into the search area. This strategy allows the feature learning of the search area to be aware of the template while keeping the individual characteristics of the template intact. To enable the network to easily preserve the information learned at different stages and ease the optimization, for the search area, we densely connect the initial input sparse pillars and the output of each stage to all subsequent stages and the target localization network, which converts pillars to bird's eye view (BEV) feature maps and predicts the state of the target with a small densely connected convolution network. Deep supervision is added to each stage to further boost the performance as well. The proposed algorithm is evaluated on the popular KITTI, nuScenes, and Waymo datasets, and the experimental results show that our method achieves promising performance compared with the state-of-the-art. Ablation study that shows the effectiveness of each component is provided as well. Code is available at https://github.com/liangp/MCSTN-3DSOT., Comment: Preprint version for IEEE Robotics and Automation Letters (RAL)

This study investigates the use of orbital angular momentum (OAM) to enhance phase estimation in Mach-Zehnder interferometers (MZIs) by employing non-Gaussian states as input resources in the presence of noise. Our research demonstrates that non-Gaussian states, particularly the photonsubtraction-then-addition (PSA) state, exhibit the best sensitivity in the presence of symmetric noise. Additionally, higher-order of Bose operator of non-Gaussian states provide better sensitivity for symmetric noise. OAM can mitigate the deterioration of noise, making it possible to estimate small phase shifts theta close to 0. OAM enhances the resolution and sensitivity of all input states and mitigating the deterioration caused by photon loss. Additionally, OAM enhances the resolution and sensitivity of all input states, enabling the sensitivity to approach the 1/N limit even under significant photon loss (e.g.,50% symmetric photon loss). These results hold promise for enhancing the sensitivity and robustness of quantum metrology, particularly in the presence of significant photon loss., Comment: 14 pages, 18 figures

In this paper, we present Consistent4D, a novel approach for generating 4D dynamic objects from uncalibrated monocular videos. Uniquely, we cast the 360-degree dynamic object reconstruction as a 4D generation problem, eliminating the need for tedious multi-view data collection and camera calibration. This is achieved by leveraging the object-level 3D-aware image diffusion model as the primary supervision signal for training Dynamic Neural Radiance Fields (DyNeRF). Specifically, we propose a Cascade DyNeRF to facilitate stable convergence and temporal continuity under the supervision signal which is discrete along the time axis. To achieve spatial and temporal consistency, we further introduce an Interpolation-driven Consistency Loss. It is optimized by minimizing the discrepancy between rendered frames from DyNeRF and interpolated frames from a pre-trained video interpolation model. Extensive experiments show that our Consistent4D can perform competitively to prior art alternatives, opening up new possibilities for 4D dynamic object generation from monocular videos, whilst also demonstrating advantage for conventional text-to-3D generation tasks. Our project page is https://consistent4d.github.io/., Comment: Technique report. Project page: https://consistent4d.github.io/

In urban planning, land use readjustment plays a pivotal role in aligning land use configurations with the current demands for sustainable urban development. However, present-day urban planning practices face two main issues. Firstly, land use decisions are predominantly dependent on human experts. Besides, while resident engagement in urban planning can promote urban sustainability and livability, it is challenging to reconcile the diverse interests of stakeholders. To address these challenges, we introduce a Consensus-based Multi-Agent Reinforcement Learning framework for real-world land use readjustment. This framework serves participatory urban planning, allowing diverse intelligent agents as stakeholder representatives to vote for preferred land use types. Within this framework, we propose a novel consensus mechanism in reward design to optimize land utilization through collective decision making. To abstract the structure of the complex urban system, the geographic information of cities is transformed into a spatial graph structure and then processed by graph neural networks. Comprehensive experiments on both traditional top-down planning and participatory planning methods from real-world communities indicate that our computational framework enhances global benefits and accommodates diverse interests, leading to improved satisfaction across different demographic groups. By integrating Multi-Agent Reinforcement Learning, our framework ensures that participatory urban planning decisions are more dynamic and adaptive to evolving community needs and provides a robust platform for automating complex real-world urban planning processes.

In this paper, we study the weak monotonicity property of p-energy related Korevaar-Schoen norms on connected nested fractals for $1 < p < \infty$. Such property has many important applications on fractals and other metric measure spaces, such as constructing p-energies (when $p = 2$ this is basically a Dirichlet form), generalizing the classical Sobolev type inequalities and the celebrated Bourgain-Brezis-Mironescu convergence., Comment: 10 pages,1 figure

Recently, the transformer has enabled the speed-oriented trackers to approach state-of-the-art (SOTA) performance with high-speed thanks to the smaller input size or the lighter feature extraction backbone, though they still substantially lag behind their corresponding performance-oriented versions. In this paper, we demonstrate that it is possible to narrow or even close this gap while achieving high tracking speed based on the smaller input size. To this end, we non-uniformly resize the cropped image to have a smaller input size while the resolution of the area where the target is more likely to appear is higher and vice versa. This enables us to solve the dilemma of attending to a larger visual field while retaining more raw information for the target despite a smaller input size. Our formulation for the non-uniform resizing can be efficiently solved through quadratic programming (QP) and naturally integrated into most of the crop-based local trackers. Comprehensive experiments on five challenging datasets based on two kinds of transformer trackers, \ie, OSTrack and TransT, demonstrate consistent improvements over them. In particular, applying our method to the speed-oriented version of OSTrack even outperforms its performance-oriented counterpart by 0.6% AUC on TNL2K, while running 50% faster and saving over 55% MACs. Codes and models are available at https://github.com/Kou-99/ZoomTrack., Comment: 19 pages, 7 figures, Accepted by NeurIPS 2023 as a Spotlight

Due to the lack of the translational symmetry, calculating the energy spectrum of an incommensurate system has always been a theoretical challenge. Here, we propose a natural approach to generalize the energy band theory to the incommensurate systems without reliance on the commensurate approximation, thus providing a comprehensive energy spectrum theory of the incommensurate systems. Except for a truncation dependent weighting factor, the formulae of this theory are formally almost identical to that of the Bloch electrons, making it particularly suitable for complex incommensurate structures. To illustrate the application of this theory, we give three typical examples: one-dimensional bichromatic and trichromatic incommensurate potential model, as well as a moir\'{e} quasicrystal. Our theory establishes a fundamental framework for understanding the incommensurate systems., Comment: 7 pages, 3 figures

The search for room-temperature superconductors has been a long-standing goal in condensed matter physics. In this study, we investigate the electronic and geometric properties of lead apatite with and without Cu doped within the frame work of the density functional theory. Based on our calculations, we found that without the Cu doped the lead apatite shows an insulator character with flat bands straddle the Fermi level. Once we introduce the O1 vacancies, the flat bands disappear. Furthermore, we analyze the effects of Cu doping on the crystal structure and electronic band structure of the material. Our calculations reveal the presence of one-dimensional channels induced by fully occupied O1 atoms, that are only 1/4 occupied in the literature, which may play a crucial role in the realization of room-temperature superconductivity. Based on our findings, we propose a possible solution to improve the quality of superconductivity by annealing the material in an oxygen atmosphere. These results contribute to a better understanding of the unusual properties of Cu-doped lead apatite and will pave the way for further exploration of its potential as a room-temperature superconductor., Comment: 8 pages, 7 figures

The recent surge of foundation models in computer vision and natural language processing opens up perspectives in utilizing multi-modal clinical data to train large models with strong generalizability. Yet pathological image datasets often lack biomedical text annotation and enrichment. Guiding data-efficient image diagnosis from the use of biomedical text knowledge becomes a substantial interest. In this paper, we propose to Connect Image and Text Embeddings (CITE) to enhance pathological image classification. CITE injects text insights gained from language models pre-trained with a broad range of biomedical texts, leading to adapt foundation models towards pathological image understanding. Through extensive experiments on the PatchGastric stomach tumor pathological image dataset, we demonstrate that CITE achieves leading performance compared with various baselines especially when training data is scarce. CITE offers insights into leveraging in-domain text knowledge to reinforce data-efficient pathological image classification. Code is available at https://github.com/Yunkun-Zhang/CITE., Comment: Accepted to MICCAI2023