Searcharxiv⌕ Search

arXiv subjects

Peng Gao

Publications and source records attributed to Peng Gao.

At least 253 records · Page 14Linked to original sources
arXiv2023

PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning

Large-scale pre-trained models have shown promising open-world performance for both vision and language tasks. However, their transferred capacity on 3D point clouds is still limited and only constrained to the classification task. In this paper, we first collaborate CLIP and GPT to be a unified 3D open-world learner, named as PointCLIP V2, which fully unleashes their potential for zero-shot 3D classification, segmentation, and detection. To better align 3D data with the pre-trained language knowledge, PointCLIP V2 contains two key designs. For the visual end, we prompt CLIP via a shape projection module to generate more realistic depth maps, narrowing the domain gap between projected point clouds with natural images. For the textual end, we prompt the GPT model to generate 3D-specific text as the input of CLIP's textual encoder. Without any training in 3D domains, our approach significantly surpasses PointCLIP by +42.90%, +40.44%, and +28.75% accuracy on three datasets for zero-shot 3D classification. On top of that, V2 can be extended to few-shot 3D classification, zero-shot 3D part segmentation, and 3D object detection in a simple manner, demonstrating our generalization ability for unified 3D open-world learning.

cs.CV↗
arXiv2023

Less is More: Towards Efficient Few-shot 3D Semantic Segmentation via Training-free Networks

To reduce the reliance on large-scale datasets, recent works in 3D segmentation resort to few-shot learning. Current 3D few-shot semantic segmentation methods first pre-train the models on `seen' classes, and then evaluate their generalization performance on `unseen' classes. However, the prior pre-training stage not only introduces excessive time overhead, but also incurs a significant domain gap on `unseen' classes. To tackle these issues, we propose an efficient Training-free Few-shot 3D Segmentation netwrok, TFS3D, and a further training-based variant, TFS3D-T. Without any learnable parameters, TFS3D extracts dense representations by trigonometric positional encodings, and achieves comparable performance to previous training-based methods. Due to the elimination of pre-training, TFS3D can alleviate the domain gap issue and save a substantial amount of time. Building upon TFS3D, TFS3D-T only requires to train a lightweight query-support transferring attention (QUEST), which enhances the interaction between the few-shot query and support data. Experiments demonstrate TFS3D-T improves previous state-of-the-art methods by +6.93% and +17.96% mIoU respectively on S3DIS and ScanNet, while reducing the training time by -90%, indicating superior effectiveness and efficiency.

cs.CV↗
arXiv2023

Controlling the 2D magnetism of CrBr$_3$ by van der Waals stacking engineering

The manipulation of two-dimensional (2D) magnetic order is of significant importance to facilitate future 2D magnets for low-power and high-speed spintronic devices. Van der Waals stacking engineering makes promises for controllable magnetism via interlayer magnetic coupling. However, directly examining the stacking order changes accompanying magnetic order transitions at the atomic scale and preparing device-ready 2D magnets with controllable magnetic orders remain elusive. Here, we demonstrate effective control of interlayer stacking in exfoliated CrBr$_3$ via thermally assisted strain engineering. The stable interlayer ferromagnetic (FM), antiferromagnetic (AFM), and FM-AFM coexistent ground states confirmed by the magnetic circular dichroism measurements are realized. Combined with the first-principles calculations, the atomically-resolved imaging technique reveals the correlation between magnetic order and interlay stacking order in the CrBr$_3$ flakes unambiguously. A tunable exchange bias effect is obtained in the mixed phase of FM and AFM states. This work will introduce new magnetic properties by controlling the stacking order, and sequence of 2D magnets, providing ample opportunities for their application in spintronic devices.

cond-mat.mtrl-sci↗
arXiv2023

Atomic-Scale Tracking Phase Transition Dynamics of Berezinskii-Kosterlitz-Thouless Polar Vortex-Antivortex

Particle-like topologies, such as vortex-antivortex (V-AV) pairs, have garnered significant attention in the field of condensed matter. However, the detailed phase transition dynamics of V-AV pairs, as exemplified by self-annihilation, motion, and dissociation, have yet to be verified in real space due to the lack of suitable experimental techniques. Here, we employ polar V-AV pairs as a model system and track their transition pathways at atomic resolution with the aid of in situ (scanning) transmission electron microscopy and phase field simulations. We demonstrate the absence of a Berezinskii-Kosterlitz-Thouless phase transition between the room-temperature quasi-long-range ordered ground phase and the high-temperature disordered phase. Instead, we observe polarization suppression in bound V-AV pairs as the temperature increases. Furthermore, electric fields can promote the vortex and antivortex to approach each other and annihilate near the interface. The elucidated intermediate dynamic behaviors of polar V-AV pairs under thermal- and electrical-fields lay the foundation for their potential applications in electronic devices. Moreover, the dynamic behaviors revealed at atomic scale provide us new insights into understanding topological phase of matter and their topological phase transitions.

physics.app-ph↗
arXiv2023

Weak and strong versions of the Kolmogorov 4/5-law for stochastic Burgers equation

For solutions of the space-periodic stochastic 1d Burgers equation we establish two versions of the Kolmogorov 4/5-law which provides an asymptotic expansion for the third moment of increments of turbulent velocity fields. We also prove for this equation an analogy of the Landau objection to possible universality of Kolmogorov's theory of turbulence, and show that the third moment is the only one which admits a universal asymptotic expansion.

math.PR↗
arXiv2023

SUG: Single-dataset Unified Generalization for 3D Point Cloud Classification

Although Domain Generalization (DG) problem has been fast-growing in the 2D image tasks, its exploration on 3D point cloud data is still insufficient and challenged by more complex and uncertain cross-domain variances with uneven inter-class modality distribution. In this paper, different from previous 2D DG works, we focus on the 3D DG problem and propose a Single-dataset Unified Generalization (SUG) framework that only leverages a single source dataset to alleviate the unforeseen domain differences faced by a well-trained source model. Specifically, we first design a Multi-grained Sub-domain Alignment (MSA) method, which can constrain the learned representations to be domain-agnostic and discriminative, by performing a multi-grained feature alignment process between the splitted sub-domains from the single source dataset. Then, a Sample-level Domain-aware Attention (SDA) strategy is presented, which can selectively enhance easy-to-adapt samples from different sub-domains according to the sample-level inter-domain distance to avoid the negative transfer. Experiments demonstrate that our SUG can boost the generalization ability for unseen target domains, even outperforming the existing unsupervised domain adaptation methods that have to access extensive target domain data. Our code is available at https://github.com/SiyuanHuang95/SUG.

cs.CV↗
arXiv2023

Reply to "Comment on newly found Charge Density Waves in infinite layer Nickelates''

Charge density waves (CDW) have been reported in NdNiO$_2$ and LaNiO$_2$ thin films grown on SrTiO$_3$ substrates using Ni-$L_3$ resonant x-ray scattering in Refs. [1-3]. In their comment [arXiv:2306.15086] on these reports, Pelliciari et al. found no evidence for a CDW in a NdNiO$_2$ film by performing fixed-momentum energy-dependent measurements. Instead, they observed a nearby non-resonant scattering peak, attributed to the (101) substrate reflection, made accessible at Ni-$L_3$ due to third harmonic light contamination. Here we present fixed-momentum energy-dependent resonant inelastic x-ray scattering measurements across Ni-$L_3$ on NdNiO$_2$, used in the preceding study [1]. We see intrinsic Ni-$L_3$ energy profiles at all measured \textbf{Q} values, including a strong resonance effect at $\mathbf{Q}_\mathrm{CDW} = (-1/3, 0, 0.316)$ reciprocal lattice units. Attempts to measure the (101) substrate peak using third harmonic light at Ni-$L_3$ at I21, Diamond were unfruitful. Our results clearly demonstrate the electronic origin of the scattering peak published in Ref. [1] and lack of a detectable structural component in the peak.

cond-mat.str-el↗
arXiv2023

ClueReader: Heterogeneous Graph Attention Network for Multi-hop Machine Reading Comprehension

Multi-hop machine reading comprehension is a challenging task in natural language processing as it requires more reasoning ability across multiple documents. Spectral models based on graph convolutional networks have shown good inferring abilities and lead to competitive results. However, the analysis and reasoning of some are inconsistent with those of humans. Inspired by the concept of grandmother cells in cognitive neuroscience, we propose a heterogeneous graph attention network model named ClueReader to imitate the grandmother cell concept. The model is designed to assemble the semantic features in multi-level representations and automatically concentrate or alleviate information for reasoning through the attention mechanism. The name ClueReader is a metaphor for the pattern of the model: it regards the subjects of queries as the starting points of clues, takes the reasoning entities as bridge points, considers the latent candidate entities as grandmother cells, and the clues end up in candidate entities. The proposed model enables the visualization of the reasoning graph, making it possible to analyze the importance of edges connecting entities and the selectivity in the mention and candidate nodes, which is easier to comprehend empirically. Evaluations on the open-domain multi-hop reading dataset WikiHop and drug-drug interaction dataset MedHop proved the validity of ClueReader and showed the feasibility of its application of the model in the molecular biology domain.

cs.CL↗
arXiv2023

Harnessing dislocation motion using an electric field

Dislocations, line defects in crystalline materials, play an essential role in the mechanical[1,2], electrical[3], optical[4], thermal[5], and phase transition[6] properties of these materials. Dislocation motion, an important mechanism underlying crystal plasticity, is critical for the hardening, processing, and application of a wide range of structural and functional materials[1,7,8]. For decades, the movement of dislocations has been widely observed in crystalline solids under mechanical loading[9-11]. However, the goal of manipulating dislocation motion via a non-mechanical field alone remains elusive. Here, we present real-time observations of dislocation motion controlled solely by an external electric field in single-crystalline zinc sulfide (ZnS). We find that 30° partial dislocations can move back and forth depending on the direction of the electric field, while 90° partial dislocations are motionless. We reveal the nonstoichiometric nature of dislocation cores using atomistic imaging and determine their charge characteristics by density functional theory calculations. The glide barriers of charged 30° partial dislocations, which are lower than those of 90° partial dislocations, further decrease under an electric field, explaining the experimental observations. This study provides direct evidence of dislocation dynamics under a non-mechanical stimulus and opens up the possibility of modulating dislocation-related properties.

cond-mat.mtrl-sci↗
arXiv2023

Enhanced Neural Beamformer with Spatial Information for Target Speech Extraction

Recently, deep learning-based beamforming algorithms have shown promising performance in target speech extraction tasks. However, most systems do not fully utilize spatial information. In this paper, we propose a target speech extraction network that utilizes spatial information to enhance the performance of neural beamformer. To achieve this, we first use the UNet-TCN structure to model input features and improve the estimation accuracy of the speech pre-separation module by avoiding information loss caused by direct dimensionality reduction in other models. Furthermore, we introduce a multi-head cross-attention mechanism that enhances the neural beamformer's perception of spatial information by making full use of the spatial information received by the array. Experimental results demonstrate that our approach, which incorporates a more reasonable target mask estimation network and a spatial information-based cross-attention mechanism into the neural beamformer, effectively improves speech separation performance.

cs.SD↗
arXiv2023

LVLM-eHub: A Comprehensive Evaluation Benchmark for Large Vision-Language Models

Large Vision-Language Models (LVLMs) have recently played a dominant role in multimodal vision-language learning. Despite the great success, it lacks a holistic evaluation of their efficacy. This paper presents a comprehensive evaluation of publicly available large multimodal models by building a LVLM evaluation Hub (LVLM-eHub). Our LVLM-eHub consists of $8$ representative LVLMs such as InstructBLIP and MiniGPT-4, which are thoroughly evaluated by a quantitative capability evaluation and an online arena platform. The former evaluates $6$ categories of multimodal capabilities of LVLMs such as visual question answering and embodied artificial intelligence on $47$ standard text-related visual benchmarks, while the latter provides the user-level evaluation of LVLMs in an open-world question-answering scenario. The study reveals several innovative findings. First, instruction-tuned LVLM with massive in-domain data such as InstructBLIP heavily overfits many existing tasks, generalizing poorly in the open-world scenario. Second, instruction-tuned LVLM with moderate instruction-following data may result in object hallucination issues (i.e., generate objects that are inconsistent with target images in the descriptions). It either makes the current evaluation metric such as CIDEr for image captioning ineffective or generates wrong answers. Third, employing a multi-turn reasoning evaluation framework can mitigate the issue of object hallucination, shedding light on developing an effective pipeline for LVLM evaluation. The findings provide a foundational framework for the conception and assessment of innovative strategies aimed at enhancing zero-shot multimodal techniques. Our LVLM-eHub will be available at https://github.com/OpenGVLab/Multi-Modality-Arena

cs.CV↗
arXiv2023

UniFormer: Unifying Convolution and Self-attention for Visual Recognition

It is a challenging task to learn discriminative representation from images and videos, due to large local redundancy and complex global dependency in these visual data. Convolution neural networks (CNNs) and vision transformers (ViTs) have been two dominant frameworks in the past few years. Though CNNs can efficiently decrease local redundancy by convolution within a small neighborhood, the limited receptive field makes it hard to capture global dependency. Alternatively, ViTs can effectively capture long-range dependency via self-attention, while blind similarity comparisons among all the tokens lead to high redundancy. To resolve these problems, we propose a novel Unified transFormer (UniFormer), which can seamlessly integrate the merits of convolution and self-attention in a concise transformer format. Different from the typical transformer blocks, the relation aggregators in our UniFormer block are equipped with local and global token affinity respectively in shallow and deep layers, allowing to tackle both redundancy and dependency for efficient and effective representation learning. Finally, we flexibly stack our UniFormer blocks into a new powerful backbone, and adopt it for various vision tasks from image to video domain, from classification to dense prediction. Without any extra training data, our UniFormer achieves 86.3 top-1 accuracy on ImageNet-1K classification. With only ImageNet-1K pre-training, it can simply achieve state-of-the-art performance in a broad range of downstream tasks, e.g., it obtains 82.9/84.8 top-1 accuracy on Kinetics-400/600, 60.9/71.2 top-1 accuracy on Sth-Sth V1/V2 video classification, 53.8 box AP and 46.4 mask AP on COCO object detection, 50.8 mIoU on ADE20K semantic segmentation, and 77.4 AP on COCO pose estimation. We further build an efficient UniFormer with 2-4x higher throughput. Code is available at https://github.com/Sense-X/UniFormer.

cs.CV↗
arXiv2023

Instruct2Act: Mapping Multi-modality Instructions to Robotic Actions with Large Language Model

Foundation models have made significant strides in various applications, including text-to-image generation, panoptic segmentation, and natural language processing. This paper presents Instruct2Act, a framework that utilizes Large Language Models to map multi-modal instructions to sequential actions for robotic manipulation tasks. Specifically, Instruct2Act employs the LLM model to generate Python programs that constitute a comprehensive perception, planning, and action loop for robotic tasks. In the perception section, pre-defined APIs are used to access multiple foundation models where the Segment Anything Model (SAM) accurately locates candidate objects, and CLIP classifies them. In this way, the framework leverages the expertise of foundation models and robotic abilities to convert complex high-level instructions into precise policy codes. Our approach is adjustable and flexible in accommodating various instruction modalities and input types and catering to specific task demands. We validated the practicality and efficiency of our approach by assessing it on robotic tasks in different scenarios within tabletop manipulation domains. Furthermore, our zero-shot method outperformed many state-of-the-art learning-based policies in several tasks. The code for our proposed approach is available at https://github.com/OpenGVLab/Instruct2Act, serving as a robust benchmark for high-level robotic instruction tasks with assorted modality inputs.

cs.RO↗
arXiv2023

Parameter is Not All You Need: Starting from Non-Parametric Networks for 3D Point Cloud Analysis

We present a Non-parametric Network for 3D point cloud analysis, Point-NN, which consists of purely non-learnable components: farthest point sampling (FPS), k-nearest neighbors (k-NN), and pooling operations, with trigonometric functions. Surprisingly, it performs well on various 3D tasks, requiring no parameters or training, and even surpasses existing fully trained models. Starting from this basic non-parametric model, we propose two extensions. First, Point-NN can serve as a base architectural framework to construct Parametric Networks by simply inserting linear layers on top. Given the superior non-parametric foundation, the derived Point-PN exhibits a high performance-efficiency trade-off with only a few learnable parameters. Second, Point-NN can be regarded as a plug-and-play module for the already trained 3D models during inference. Point-NN captures the complementary geometric knowledge and enhances existing methods for different 3D benchmarks without re-training. We hope our work may cast a light on the community for understanding 3D point clouds with non-parametric methods. Code is available at https://github.com/ZrrSkywalker/Point-NN.

cs.CV↗
arXiv2023

LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model

How to efficiently transform large language models (LLMs) into instruction followers is recently a popular research direction, while training LLM for multi-modal reasoning remains less explored. Although the recent LLaMA-Adapter demonstrates the potential to handle visual inputs with LLMs, it still cannot generalize well to open-ended visual instructions and lags behind GPT-4. In this paper, we present LLaMA-Adapter V2, a parameter-efficient visual instruction model. Specifically, we first augment LLaMA-Adapter by unlocking more learnable parameters (e.g., norm, bias and scale), which distribute the instruction-following ability across the entire LLaMA model besides adapters. Secondly, we propose an early fusion strategy to feed visual tokens only into the early LLM layers, contributing to better visual knowledge incorporation. Thirdly, a joint training paradigm of image-text pairs and instruction-following data is introduced by optimizing disjoint groups of learnable parameters. This strategy effectively alleviates the interference between the two tasks of image-text alignment and instruction following and achieves strong multi-modal reasoning with only a small-scale image-text and instruction dataset. During inference, we incorporate additional expert models (e.g. captioning/OCR systems) into LLaMA-Adapter to further enhance its image understanding capability without incurring training costs. Compared to the original LLaMA-Adapter, our LLaMA-Adapter V2 can perform open-ended multi-modal instructions by merely introducing 14M parameters over LLaMA. The newly designed framework also exhibits stronger language-only instruction-following capabilities and even excels in chat interactions. Our code and models are available at https://github.com/ZrrSkywalker/LLaMA-Adapter.

cs.CV↗
arXiv2023

Ratios conjecture for quadratic twist of modular $L$-functions

We develop the $L$-functions ratios conjecture with one shift in the numerator and denominator in certain ranges for the family of quadratic twist of modular $L$-functions using multiple Dirichlet series under the generalized Riemann hypothesis.

math.NT↗
arXiv2023

Filter Pruning via Filters Similarity in Consecutive Layers

Filter pruning is widely adopted to compress and accelerate the Convolutional Neural Networks (CNNs), but most previous works ignore the relationship between filters and channels in different layers. Processing each layer independently fails to utilize the collaborative relationship across layers. In this paper, we intuitively propose a novel pruning method by explicitly leveraging the Filters Similarity in Consecutive Layers (FSCL). FSCL compresses models by pruning filters whose corresponding features are more worthless in the model. The extensive experiments demonstrate the effectiveness of FSCL, and it yields remarkable improvement over state-of-the-art on accuracy, FLOPs and parameter reduction on several benchmark models and datasets.

cs.CV↗