arXiv subjects

Xu Yang

Publications and source records attributed to Xu Yang.

At least 289 records · Page 16Linked to original sources

arXiv2023

Transforming Visual Scene Graphs to Image Captions

We propose to Transform Scene Graphs (TSG) into more descriptive captions. In TSG, we apply multi-head attention (MHA) to design the Graph Neural Network (GNN) for embedding scene graphs. After embedding, different graph embeddings contain diverse specific knowledge for generating the words with different part-of-speech, e.g., object/attribute embedding is good for generating nouns/adjectives. Motivated by this, we design a Mixture-of-Expert (MOE)-based decoder, where each expert is built on MHA, for discriminating the graph embeddings to generate different kinds of words. Since both the encoder and decoder are built based on the MHA, as a result, we construct a homogeneous encoder-decoder unlike the previous heterogeneous ones which usually apply Fully-Connected-based GNN and LSTM-based decoder. The homogeneous architecture enables us to unify the training configuration of the whole model instead of specifying different training strategies for diverse sub-networks as in the heterogeneous pipeline, which releases the training difficulty. Extensive experiments on the MS-COCO captioning benchmark validate the effectiveness of our TSG. The code is in: https://github.com/GaryJiajia/TSG.

cs.CV↗

arXiv2023

Manipulating the Label Space for In-Context Classification

After pre-training by generating the next word conditional on previous words, the Language Model (LM) acquires the ability of In-Context Learning (ICL) that can learn a new task conditional on the context of the given in-context examples (ICEs). Similarly, visually-conditioned Language Modelling is also used to train Vision-Language Models (VLMs) with ICL ability. However, such VLMs typically exhibit weaker classification abilities compared to contrastive learning-based models like CLIP, since the Language Modelling objective does not directly contrast whether an object is paired with a text. To improve the ICL of classification, using more ICEs to provide more knowledge is a straightforward way. However, this may largely increase the selection time, and more importantly, the inclusion of additional in-context images tends to extend the length of the in-context sequence beyond the processing capacity of a VLM. To alleviate these limitations, we propose to manipulate the label space of each ICE to increase its knowledge density, allowing for fewer ICEs to convey as much information as a larger set would. Specifically, we propose two strategies which are Label Distribution Enhancement and Visual Descriptions Enhancement to improve In-context classification performance on diverse datasets, including the classic ImageNet and more fine-grained datasets like CUB-200. Specifically, using our approach on ImageNet, we increase accuracy from 74.70\% in a 4-shot setting to 76.21\% with just 2 shots. surpassing CLIP by 0.67\%. On CUB-200, our method raises 1-shot accuracy from 48.86\% to 69.05\%, 12.15\% higher than CLIP. The code is given in https://anonymous.4open.science/r/MLS_ICC.

cs.CV↗

arXiv2023

How to Configure Good In-Context Sequence for Visual Question Answering

Inspired by the success of Large Language Models in dealing with new tasks via In-Context Learning (ICL) in NLP, researchers have also developed Large Vision-Language Models (LVLMs) with ICL capabilities. However, when implementing ICL using these LVLMs, researchers usually resort to the simplest way like random sampling to configure the in-context sequence, thus leading to sub-optimal results. To enhance the ICL performance, in this study, we use Visual Question Answering (VQA) as case study to explore diverse in-context configurations to find the powerful ones. Additionally, through observing the changes of the LVLM outputs by altering the in-context sequence, we gain insights into the inner properties of LVLMs, improving our understanding of them. Specifically, to explore in-context configurations, we design diverse retrieval methods and employ different strategies to manipulate the retrieved demonstrations. Through exhaustive experiments on three VQA datasets: VQAv2, VizWiz, and OK-VQA, we uncover three important inner properties of the applied LVLM and demonstrate which strategies can consistently improve the ICL VQA performance. Our code is provided in: https://github.com/GaryJiajia/OFv2_ICL_VQA.

cs.CV↗

arXiv2023

Deconfounded Image Captioning: A Causal Retrospect

Dataset bias in vision-language tasks is becoming one of the main problems which hinders the progress of our community. Existing solutions lack a principled analysis about why modern image captioners easily collapse into dataset bias. In this paper, we present a novel perspective: Deconfounded Image Captioning (DIC), to find out the answer of this question, then retrospect modern neural image captioners, and finally propose a DIC framework: DICv1.0 to alleviate the negative effects brought by dataset bias. DIC is based on causal inference, whose two principles: the backdoor and front-door adjustments, help us review previous studies and design new effective models. In particular, we showcase that DICv1.0 can strengthen two prevailing captioning models and can achieve a single-model 131.1 CIDEr-D and 128.4 c40 CIDEr-D on Karpathy split and online split of the challenging MS COCO dataset, respectively. Interestingly, DICv1.0 is a natural derivation from our causal retrospect, which opens promising directions for image captioning.

cs.CV↗

arXiv2023

ChartLlama: A Multimodal LLM for Chart Understanding and Generation

Multi-modal large language models have demonstrated impressive performances on most vision-language tasks. However, the model generally lacks the understanding capabilities for specific domain data, particularly when it comes to interpreting chart figures. This is mainly due to the lack of relevant multi-modal instruction tuning datasets. In this article, we create a high-quality instruction-tuning dataset leveraging GPT-4. We develop a multi-step data generation process in which different steps are responsible for generating tabular data, creating chart figures, and designing instruction tuning data separately. Our method's flexibility enables us to generate diverse, high-quality instruction-tuning data consistently and efficiently while maintaining a low resource expenditure. Additionally, it allows us to incorporate a wider variety of chart and task types not yet featured in existing datasets. Next, we introduce ChartLlama, a multi-modal large language model that we've trained using our created dataset. ChartLlama outperforms all prior methods in ChartQA, Chart-to-text, and Chart-extraction evaluation benchmarks. Additionally, ChartLlama significantly improves upon the baseline in our specially compiled chart dataset, which includes new chart and task types. The results of ChartLlama confirm the value and huge potential of our proposed data generation method in enhancing chart comprehension.

cs.CV↗

arXiv2023

Rethinking Residual Connection in Training Large-Scale Spiking Neural Networks

Spiking Neural Network (SNN) is known as the most famous brain-inspired model, but the non-differentiable spiking mechanism makes it hard to train large-scale SNNs. To facilitate the training of large-scale SNNs, many training methods are borrowed from Artificial Neural Networks (ANNs), among which deep residual learning is the most commonly used. But the unique features of SNNs make prior intuition built upon ANNs not available for SNNs. Although there are a few studies that have made some pioneer attempts on the topology of Spiking ResNet, the advantages of different connections remain unclear. To tackle this issue, we analyze the merits and limitations of various residual connections and empirically demonstrate our ideas with extensive experiments. Then, based on our observations, we abstract the best-performing connections into densely additive (DA) connection, extend such a concept to other topologies, and propose four architectures for training large-scale SNNs, termed DANet, which brings up to 13.24% accuracy gain on ImageNet. Besides, in order to present a detailed methodology for designing the topology of large-scale SNNs, we further conduct in-depth discussions on their applicable scenarios in terms of their performance on various scales of datasets and demonstrate their advantages over prior architectures. At a low training expense, our best-performing ResNet-50/101/152 obtain 73.71%/76.13%/77.22% top-1 accuracy on ImageNet with 4 time steps. We believe that this work shall give more insights for future works to design the topology of their networks and promote the development of large-scale SNNs. The code will be publicly available.

cs.NE↗

arXiv2023

Rethinking Evaluation Metrics of Open-Vocabulary Segmentaion

In this paper, we highlight a problem of evaluation metrics adopted in the open-vocabulary segmentation. That is, the evaluation process still heavily relies on closed-set metrics on zero-shot or cross-dataset pipelines without considering the similarity between predicted and ground truth categories. To tackle this issue, we first survey eleven similarity measurements between two categorical words using WordNet linguistics statistics, text embedding, and language models by comprehensive quantitative analysis and user study. Built upon those explored measurements, we designed novel evaluation metrics, namely Open mIoU, Open AP, and Open PQ, tailored for three open-vocabulary segmentation tasks. We benchmarked the proposed evaluation metrics on 12 open-vocabulary methods of three segmentation tasks. Even though the relative subjectivity of similarity distance, we demonstrate that our metrics can still well evaluate the open ability of the existing open-vocabulary segmentation methods. We hope that our work can bring with the community new thinking about how to evaluate the open ability of models. The evaluation code is released in github.

cs.CV↗

arXiv2023

Genes in Intelligent Agents

The genes in nature give the lives on earth the current biological intelligence through transmission and accumulation over billions of years. Inspired by the biological intelligence, artificial intelligence (AI) has devoted to building the machine intelligence. Although it has achieved thriving successes, the machine intelligence still lags far behind the biological intelligence. The reason may lie in that animals are born with some intelligence encoded in their genes, but machines lack such intelligence and learn from scratch. Inspired by the genes of animals, we define the ``genes'' of machines named as the ``learngenes'' and propose the Genetic Reinforcement Learning (GRL). GRL is a computational framework that simulates the evolution of organisms in reinforcement learning (RL) and leverages the learngenes to learn and evolve the intelligence agents. Leveraging GRL, we first show that the learngenes take the form of the fragments of the agents' neural networks and can be inherited across generations. Second, we validate that the learngenes can transfer ancestral experience to the agents and bring them instincts and strong learning abilities. Third, we justify the Lamarckian inheritance of the intelligent agents and the continuous evolution of the learngenes. Overall, the learngenes have taken the machine intelligence one more step toward the biological intelligence.

cs.NE↗

arXiv2023

Leveraging Large Language Model for Automatic Evolving of Industrial Data-Centric R&D Cycle

In the wake of relentless digital transformation, data-driven solutions are emerging as powerful tools to address multifarious industrial tasks such as forecasting, anomaly detection, planning, and even complex decision-making. Although data-centric R&D has been pivotal in harnessing these solutions, it often comes with significant costs in terms of human, computational, and time resources. This paper delves into the potential of large language models (LLMs) to expedite the evolution cycle of data-centric R&D. Assessing the foundational elements of data-centric R&D, including heterogeneous task-related data, multi-facet domain knowledge, and diverse computing-functional tools, we explore how well LLMs can understand domain-specific requirements, generate professional ideas, utilize domain-specific tools to conduct experiments, interpret results, and incorporate knowledge from past endeavors to tackle new challenges. We take quantitative investment research as a typical example of industrial data-centric R&D scenario and verified our proposed framework upon our full-stack open-sourced quantitative research platform Qlib and obtained promising results which shed light on our vision of automatic evolving of industrial data-centric R&D cycle.

cs.AI↗

arXiv2023

More on Rainbow Cliques in Edge-Colored Graphs

In an edge-colored graph $G$, a rainbow clique $K_k$ is a $k$-complete subgraph in which all the edges have distinct colors. Let $e(G)$ and $c(G)$ be the number of edges and colors in $G$, respectively. In this paper, we show that for any $\varepsilon>0$, if $e(G)+c(G) \geq (1+\frac{k-3}{k-2}+2\varepsilon) {n\choose 2}$ and $k\geq 3$, then for sufficiently large $n$, the number of rainbow cliques $K_k$ in $G$ is $Ω(n^k)$. We also characterize the extremal graphs $G$ without a rainbow clique $K_k$, for $k=4,5$, when $e(G)+c(G)$ is maximum. Our results not only address existing questions but also complete the findings of Ehard and Mohr (Ehard and Mohr, Rainbow triangles and cliques in edge-colored graphs. {\it European Journal of Combinatorics, 84:103037,2020}).

math.CO↗

arXiv2023

Learngene: Inheriting Condensed Knowledge from the Ancestry Model to Descendant Models

During the continuous evolution of one organism's ancestry, its genes accumulate extensive experiences and knowledge, enabling newborn descendants to rapidly adapt to their specific environments. Motivated by this observation, we propose a novel machine learning paradigm Learngene to enable learning models to incorporate three key characteristics of genes. (i) Accumulating: the knowledge is accumulated during the continuous learning of an ancestry model. (ii) Condensing: the extensive accumulated knowledge is condensed into a much more compact information piece, i.e., learngene. (iii) Inheriting: the condensed learngene is inherited to make it easier for descendant models to adapt to new environments. Since accumulating has been studied in well-established paradigms like large-scale pre-training and lifelong learning, we focus on condensing and inheriting, which induces three key issues and we provide the preliminary solutions to these issues in this paper: (i) Learngene Form: the learngene is set to a few integral layers that can preserve significance. (ii) Learngene Condensing: we identify which layers among the ancestry model have the most similarity as one pseudo descendant model. (iii) Learngene Inheriting: to construct distinct descendant models for the specific downstream tasks, we stack some randomly initialized layers to the learngene layers. Extensive experiments across various settings, including using different network architectures like Vision Transformer (ViT) and Convolutional Neural Networks (CNNs) on different datasets, are carried out to confirm four advantages of Learngene: it makes the descendant models 1) converge more quickly, 2) exhibit less sensitivity to hyperparameters, 3) perform better, and 4) require fewer training samples to converge.

cs.LG↗

arXiv2023

On $p$-Dirac Equation on Compact Spin Manifolds

By using the Ljusternik-Schnirelman principle, we establish the existence of a nondecreasing sequence of nonnegative eigenvalues for the p-Dirac operator on compact spin manifold. Using the biorthogonal system theory on separable Banach space and some critical point theorems, we prove the existence and multiplicity of solutions to p-superlinear and p-sublinear nonlinear p-Dirac equations on compact spin manifold.

math.AP↗

arXiv2023

Directional router and controllable non-reciprocity transmission based on phase and pathway coherence

A multi-channel quantum router with four nodal cavities is constructed by two coupled-resonator waveguides and four single cavities. We can achieve directional routing by adjusting the probability of photon exiting from the specified port to close to 100% based on multiple pathways between the photon from the incident port to the outgoing port in this hybrid system. Under the effect of phase difference between two classical light fields, the mutual interference between different pathways can be adjusted to destructive interference or constructive interference, which lays the foundation for the increase and decrease of the routing probability. The influence of different parameter values on single photon routing probability is also studied. By studying the analytic formula of probability amplitude, we get the physical mechanism of exiting ports being closed under certain parameter conditions and the phase relationship between the backward transmission and the original direction transmission of photons. Furthermore the non-reciprocal transmission and directional routing beyond chiral coupling can also be realized, which provides new possibilities for the study of quantum routers and new insights for the study of photon transmission characteristics.

quant-ph↗

arXiv2023

Learning to Collocate Visual-Linguistic Neural Modules for Image Captioning

Humans tend to decompose a sentence into different parts like \textsc{sth do sth at someplace} and then fill each part with certain content. Inspired by this, we follow the \textit{principle of modular design} to propose a novel image captioner: learning to Collocate Visual-Linguistic Neural Modules (CVLNM). Unlike the \re{widely used} neural module networks in VQA, where the language (\ie, question) is fully observable, \re{the task of collocating visual-linguistic modules is more challenging.} This is because the language is only partially observable, for which we need to dynamically collocate the modules during the process of image captioning. To sum up, we make the following technical contributions to design and train our CVLNM: 1) \textit{distinguishable module design} -- \re{four modules in the encoder} including one linguistic module for function words and three visual modules for different content words (\ie, noun, adjective, and verb) and another linguistic one in the decoder for commonsense reasoning, 2) a self-attention based \textit{module controller} for robustifying the visual reasoning, 3) a part-of-speech based \textit{syntax loss} imposed on the module controller for further regularizing the training of our CVLNM. Extensive experiments on the MS-COCO dataset show that our CVLNM is more effective, \eg, achieving a new state-of-the-art 129.5 CIDEr-D, and more robust, \eg, being less likely to overfit to dataset bias and suffering less when fewer training samples are available. Codes are available at \url{https://github.com/GCYZSL/CVLMN}

cs.CV↗

arXiv2023

SC-ML: Self-supervised Counterfactual Metric Learning for Debiased Visual Question Answering

Visual question answering (VQA) is a critical multimodal task in which an agent must answer questions according to the visual cue. Unfortunately, language bias is a common problem in VQA, which refers to the model generating answers only by associating with the questions while ignoring the visual content, resulting in biased results. We tackle the language bias problem by proposing a self-supervised counterfactual metric learning (SC-ML) method to focus the image features better. SC-ML can adaptively select the question-relevant visual features to answer the question, reducing the negative influence of question-irrelevant visual features on inferring answers. In addition, question-irrelevant visual features can be seamlessly incorporated into counterfactual training schemes to further boost robustness. Extensive experiments have proved the effectiveness of our method with improved results on the VQA-CP dataset. Our code will be made publicly available.

cs.CV↗

arXiv2023

Spatial Attention and Syntax Rule Enhanced Tree Decoder for Offine Handwritten Mathematical Expression Recognition

Offline Handwritten Mathematical Expression Recognition (HMER) has been dramatically advanced recently by employing tree decoders as part of the encoder-decoder method. Despite the tree decoder-based methods regard the expressions as a tree and parse 2D spatial structure to the tree nodes sequence, the performance of existing works is still poor due to the inevitable tree nodes prediction errors. Besides, they lack syntax rules to regulate the output of expressions. In this paper, we propose a novel model called Spatial Attention and Syntax Rule Enhanced Tree Decoder (SS-TD), which is equipped with spatial attention mechanism to alleviate the prediction error of tree structure and use syntax masks (obtained from the transformation of syntax rules) to constrain the occurrence of ungrammatical mathematical expression. In this way, our model can effectively describe tree structure and increase the accuracy of output expression. Experiments show that SS-TD achieves better recognition performance than prior models on CROHME 14/16/19 datasets, demonstrating the effectiveness of our model.

cs.CV↗

arXiv2023

Well-posedness of the martingale problem for super-Brownian motion with interactive branching

In this paper a martingale problem for super-Brownian motion with interactive branching is derived. The uniqueness of the solution to the martingale problem is obtained by using the pathwise uniqueness of the solution to a corresponding system of SPDEs with proper boundary conditions. The existence of the solution to the martingale problem and the local Hölder continuity of the density process are also studied.

math.PR↗

arXiv2023

Learning Trajectory-Word Alignments for Video-Language Tasks

In a video, an object usually appears as the trajectory, i.e., it spans over a few spatial but longer temporal patches, that contains abundant spatiotemporal contexts. However, modern Video-Language BERTs (VDL-BERTs) neglect this trajectory characteristic that they usually follow image-language BERTs (IL-BERTs) to deploy the patch-to-word (P2W) attention that may over-exploit trivial spatial contexts and neglect significant temporal contexts. To amend this, we propose a novel TW-BERT to learn Trajectory-Word alignment by a newly designed trajectory-to-word (T2W) attention for solving video-language tasks. Moreover, previous VDL-BERTs usually uniformly sample a few frames into the model while different trajectories have diverse graininess, i.e., some trajectories span longer frames and some span shorter, and using a few frames will lose certain useful temporal contexts. However, simply sampling more frames will also make pre-training infeasible due to the largely increased training burdens. To alleviate the problem, during the fine-tuning stage, we insert a novel Hierarchical Frame-Selector (HFS) module into the video encoder. HFS gradually selects the suitable frames conditioned on the text context for the later cross-modal encoder to learn better trajectory-word alignments. By the proposed T2W attention and HFS, our TW-BERT achieves SOTA performances on text-to-video retrieval tasks, and comparable performances on video question-answering tasks with some VDL-BERTs trained on much more data. The code will be available in the supplementary material.

cs.CV↗