PyTorchKR Papers

최신 논문

총 9,339편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2209.13759

Structured Summarization: Unified Text Segmentation and Segment Labeling as a Generation Task

Text segmentation aims to divide text into contiguous, semantically coherent segments, while segment labeling deals with producing labels for each segment. Past work has shown success in tackling segmentation and labeling for documents and conversations. This has been possible with a combination of task-specific pipelines, supervised and unsupervised learning objectives. In this work, we propose a single encoder-decoder neural network that can handle long documents and conversations, trained simultaneously for both segmentation and segment labeling using only standard supervision. We successfully show a way to solve the combined task as a pure generation task, which we refer to as structured summarization. We apply the same technique to both document and conversational data, and we show state of the art performance across datasets for both segmentation and labeling, under both high- and low-resource settings. Our results establish a strong case for considering text segmentation and segment labeling as a whole, and moving towards general-purpose techniques that don't depend on domain expertise or task-specific components.

arXiv:2209.14102

Segmentation method of U-net sheet metal engineering drawing based on CBAM attention mechanism

In the manufacturing process of heavy industrial equipment, the specific unit in the welding diagram is first manually redrawn and then the corresponding sheet metal parts are cut, which is inefficient. To this end, this paper proposes a U-net-based method for the segmentation and extraction of specific units in welding engineering drawings. This method enables the cutting device to automatically segment specific graphic units according to visual information and automatically cut out sheet metal parts of corresponding shapes according to the segmentation results. This process is more efficient than traditional human-assisted cutting. Two weaknesses in the U-net network will lead to a decrease in segmentation performance: first, the focus on global semantic feature information is weak, and second, there is a large dimensional difference between shallow encoder features and deep decoder features. Based on the CBAM (Convolutional Block Attention Module) attention mechanism, this paper proposes a U-net jump structure model with an attention mechanism to improve the network's global semantic feature extraction ability. In addition, a U-net attention mechanism model with dual pooling convolution fusion is designed, the deep encoder's maximum pooling + convolution features and the shallow encoder's average pooling + convolution features are fused vertically to reduce the dimension difference between the shallow encoder and deep decoder. The dual-pool convolutional attention jump structure replaces the traditional U-net jump structure, which can effectively improve the specific unit segmentation performance of the welding engineering drawing. Using vgg16 as the backbone network, experiments have verified that the IoU, mAP, and Accu of our model in the welding engineering drawing dataset segmentation task are 84.72%, 86.84%, and 99.42%, respectively.

arXiv:2209.13959

Dynamic MDETR: A Dynamic Multimodal Transformer Decoder for Visual Grounding

Multimodal transformer exhibits high capacity and flexibility to align image and text for visual grounding. However, the existing encoder-only grounding framework (e.g., TransVG) suffers from heavy computation due to the self-attention operation with quadratic time complexity. To address this issue, we present a new multimodal transformer architecture, coined as Dynamic Mutilmodal DETR (Dynamic MDETR), by decoupling the whole grounding process into encoding and decoding phases. The key observation is that there exists high spatial redundancy in images. Thus, we devise a new dynamic multimodal transformer decoder by exploiting this sparsity prior to speed up the visual grounding process. Specifically, our dynamic decoder is composed of a 2D adaptive sampling module and a text guided decoding module. The sampling module aims to select these informative patches by predicting the offsets with respect to a reference point, while the decoding module works for extracting the grounded object information by performing cross attention between image features and text features. These two modules are stacked alternatively to gradually bridge the modality gap and iteratively refine the reference point of grounded object, eventually realizing the objective of visual grounding. Extensive experiments on five benchmarks demonstrate that our proposed Dynamic MDETR achieves competitive trade-offs between computation and accuracy. Notably, using only 9% feature points in the decoder, we can reduce ~44% GFLOPs of the multimodal transformer, but still get higher accuracy than the encoder-only counterpart. In addition, to verify its generalization ability and scale up our Dynamic MDETR, we build the first one-stage CLIP empowered visual grounding framework, and achieve the state-of-the-art performance on these benchmarks.

arXiv:2209.13533

Denoising Diffusion Error Correction Codes

Error correction code (ECC) is an integral part of the physical communication layer, ensuring reliable data transfer over noisy channels. Recently, neural decoders have demonstrated their advantage over classical decoding techniques. However, recent state-of-the-art neural decoders suffer from high complexity and lack the important iterative scheme characteristic of many legacy decoders. In this work, we propose to employ denoising diffusion models for the soft decoding of linear codes at arbitrary block lengths. Our framework models the forward channel corruption as a series of diffusion steps that can be reversed iteratively. Three contributions are made: (i) a diffusion process suitable for the decoding setting is introduced, (ii) the neural diffusion decoder is conditioned on the number of parity errors, which indicates the level of corruption at a given step, (iii) a line search procedure based on the code's syndrome obtains the optimal reverse diffusion step size. The proposed approach demonstrates the power of diffusion models for ECC and is able to achieve state of the art accuracy, outperforming the other neural decoders by sizable margins, even for a single reverse diffusion step.

arXiv:2209.13023공식 코드 · Papers with Code

Lex2Sent: A bagging approach to unsupervised sentiment analysis

Unsupervised text classification, with its most common form being sentiment analysis, used to be performed by counting words in a text that were stored in a lexicon, which assigns each word to one class or as a neutral word. In recent years, these lexicon-based methods fell out of favor and were replaced by computationally demanding fine-tuning techniques for encoder-only models such as BERT and zero-shot classification using decoder-only models such as GPT-4. In this paper, we propose an alternative approach: Lex2Sent, which provides improvement over classic lexicon methods but does not require any GPU or external hardware. To classify texts, we train embedding models to determine the distances between document embeddings and the embeddings of the parts of a suitable lexicon. We employ resampling, which results in a bagging effect, boosting the performance of the classification. We show that our model outperforms lexica and provides a basis for a high performing few-shot fine-tuning approach in the task of binary sentiment analysis.

arXiv:2209.13171

RepsNet: Combining Vision with Language for Automated Medical Reports

Writing reports by analyzing medical images is error-prone for inexperienced practitioners and time consuming for experienced ones. In this work, we present RepsNet that adapts pre-trained vision and language models to interpret medical images and generate automated reports in natural language. RepsNet consists of an encoder-decoder model: the encoder aligns the images with natural language descriptions via contrastive learning, while the decoder predicts answers by conditioning on encoded images and prior context of descriptions retrieved by nearest neighbor search. We formulate the problem in a visual question answering setting to handle both categorical and descriptive natural language answers. We perform experiments on two challenging tasks of medical visual question answering (VQA-Rad) and report generation (IU-Xray) on radiology image datasets. Results show that RepsNet outperforms state-of-the-art methods with 81.08 % classification accuracy on VQA-Rad 2018 and 0.58 BLEU-1 score on IU-Xray. Supplementary details are available at https://sites.google.com/view/repsnet

arXiv:2209.13306공식 코드 · Papers with Code

Embracing Consistency: A One-Stage Approach for Spatio-Temporal Video Grounding

Spatio-Temporal video grounding (STVG) focuses on retrieving the spatio-temporal tube of a specific object depicted by a free-form textual expression. Existing approaches mainly treat this complicated task as a parallel frame-grounding problem and thus suffer from two types of inconsistency drawbacks: feature alignment inconsistency and prediction inconsistency. In this paper, we present an end-to-end one-stage framework, termed Spatio-Temporal Consistency-Aware Transformer (STCAT), to alleviate these issues. Specially, we introduce a novel multi-modal template as the global objective to address this task, which explicitly constricts the grounding region and associates the predictions among all video frames. Moreover, to generate the above template under sufficient video-textual perception, an encoder-decoder architecture is proposed for effective global context modeling. Thanks to these critical designs, STCAT enjoys more consistent cross-modal feature alignment and tube prediction without reliance on any pre-trained object detectors. Extensive experiments show that our method outperforms previous state-of-the-arts with clear margins on two challenging video benchmarks (VidSTG and HC-STVG), illustrating the superiority of the proposed framework to better understanding the association between vision and natural language. Code is publicly available at https://github.com/jy0205/STCAT.

arXiv:2209.13002

Automated Urban Planning aware Spatial Hierarchies and Human Instructions

Traditional urban planning demands urban experts to spend considerable time and effort producing an optimal urban plan under many architectural constraints. The remarkable imaginative ability of deep generative learning provides hope for renovating urban planning. While automated urban planners have been examined, they are constrained because of the following: 1) neglecting human requirements in urban planning; 2) omitting spatial hierarchies in urban planning, and 3) lacking numerous urban plan data samples. To overcome these limitations, we propose a novel, deep, human-instructed urban planner. In the preliminary work, we formulate it into an encoder-decoder paradigm. The encoder is to learn the information distribution of surrounding contexts, human instructions, and land-use configuration. The decoder is to reconstruct the land-use configuration and the associated urban functional zones. The reconstruction procedure will capture the spatial hierarchies between functional zones and spatial grids. Meanwhile, we introduce a variational Gaussian mechanism to mitigate the data sparsity issue. Even though early work has led to good results, the performance of generation is still unstable because the way spatial hierarchies are captured may lead to unclear optimization directions. In this journal version, we propose a cascading deep generative framework based on generative adversarial networks (GANs) to solve this problem, inspired by the workflow of urban experts. In particular, the purpose of the first GAN is to build urban functional zones based on information from human instructions and surrounding contexts. The second GAN will produce the land-use configuration based on the functional zones that have been constructed. Additionally, we provide a conditioning augmentation module to augment data samples. Finally, we conduct extensive experiments to validate the efficacy of our work.

arXiv:2209.12733

Informative Text Generation from Knowledge Triples

As the development of the encoder-decoder architecture, researchers are able to study the text generation tasks with broader types of data. Among them, KB-to-text aims at converting a set of knowledge triples into human readable sentences. In the original setting, the task assumes that the input triples and the text are exactly aligned in the perspective of the embodied knowledge/information. In this paper, we extend this setting and explore how to facilitate the trained model to generate more informative text, namely, containing more information about the triple entities but not conveyed by the input triples. To solve this problem, we propose a novel memory augmented generator that employs a memory network to memorize the useful knowledge learned during the training and utilizes such information together with the input triples to generate text in the operational or testing phase. We derive a dataset from WebNLG for our new setting and conduct extensive experiments to investigate the effectiveness of our model as well as uncover the intrinsic characteristics of the setting.

arXiv:2209.12590

Learning to Drop Out: An Adversarial Approach to Training Sequence VAEs

In principle, applying variational autoencoders (VAEs) to sequential data offers a method for controlled sequence generation, manipulation, and structured representation learning. However, training sequence VAEs is challenging: autoregressive decoders can often explain the data without utilizing the latent space, known as posterior collapse. To mitigate this, state-of-the-art models weaken the powerful decoder by applying uniformly random dropout to the decoder input. We show theoretically that this removes pointwise mutual information provided by the decoder input, which is compensated for by utilizing the latent space. We then propose an adversarial training strategy to achieve information-based stochastic dropout. Compared to uniform dropout on standard text benchmark datasets, our targeted approach increases both sequence modeling performance and the information captured in the latent space.

arXiv:2209.12866공식 코드 · Papers with Code

SAPA: Similarity-Aware Point Affiliation for Feature Upsampling

We introduce point affiliation into feature upsampling, a notion that describes the affiliation of each upsampled point to a semantic cluster formed by local decoder feature points with semantic similarity. By rethinking point affiliation, we present a generic formulation for generating upsampling kernels. The kernels encourage not only semantic smoothness but also boundary sharpness in the upsampled feature maps. Such properties are particularly useful for some dense prediction tasks such as semantic segmentation. The key idea of our formulation is to generate similarity-aware kernels by comparing the similarity between each encoder feature point and the spatially associated local region of decoder features. In this way, the encoder feature point can function as a cue to inform the semantic cluster of upsampled feature points. To embody the formulation, we further instantiate a lightweight upsampling operator, termed Similarity-Aware Point Affiliation (SAPA), and investigate its variants. SAPA invites consistent performance improvements on a number of dense prediction tasks, including semantic segmentation, object detection, depth estimation, and image matting. Code is available at: https://github.com/poppinace/sapa

arXiv:2209.12643공식 코드 · Papers with Code

PiFold: Toward effective and efficient protein inverse folding

How can we design protein sequences folding into the desired structures effectively and efficiently? AI methods for structure-based protein design have attracted increasing attention in recent years; however, few methods can simultaneously improve the accuracy and efficiency due to the lack of expressive features and autoregressive sequence decoder. To address these issues, we propose PiFold, which contains a novel residue featurizer and PiGNN layers to generate protein sequences in a one-shot way with improved recovery. Experiments show that PiFold could achieve 51.66\% recovery on CATH 4.2, while the inference speed is 70 times faster than the autoregressive competitors. In addition, PiFold achieves 58.72\% and 60.42\% recovery scores on TS50 and TS500, respectively. We conduct comprehensive ablation studies to reveal the role of different types of protein features and model designs, inspiring further simplification and improvement. The PyTorch code is available at \href{https://github.com/A4Bio/PiFold}{GitHub}.

arXiv:2209.11963

A Deep Investigation of RNN and Self-attention for the Cyrillic-Traditional Mongolian Bidirectional Conversion

Cyrillic and Traditional Mongolian are the two main members of the Mongolian writing system. The Cyrillic-Traditional Mongolian Bidirectional Conversion (CTMBC) task includes two conversion processes, including Cyrillic Mongolian to Traditional Mongolian (C2T) and Traditional Mongolian to Cyrillic Mongolian conversions (T2C). Previous researchers adopted the traditional joint sequence model, since the CTMBC task is a natural Sequence-to-Sequence (Seq2Seq) modeling problem. Recent studies have shown that Recurrent Neural Network (RNN) and Self-attention (or Transformer) based encoder-decoder models have shown significant improvement in machine translation tasks between some major languages, such as Mandarin, English, French, etc. However, an open problem remains as to whether the CTMBC quality can be improved by utilizing the RNN and Transformer models. To answer this question, this paper investigates the utility of these two powerful techniques for CTMBC task combined with agglutinative characteristics of Mongolian language. We build the encoder-decoder based CTMBC model based on RNN and Transformer respectively and compare the different network configurations deeply. The experimental results show that both RNN and Transformer models outperform the traditional joint sequence model, where the Transformer achieves the best performance. Compared with the joint sequence baseline, the word error rate (WER) of the Transformer for C2T and T2C decreased by 5.72\% and 5.06\% respectively.

arXiv:2209.12164공식 코드 · Papers with Code

Multi-modal Segment Assemblage Network for Ad Video Editing with Importance-Coherence Reward

Advertisement video editing aims to automatically edit advertising videos into shorter videos while retaining coherent content and crucial information conveyed by advertisers. It mainly contains two stages: video segmentation and segment assemblage. The existing method performs well at video segmentation stages but suffers from the problems of dependencies on extra cumbersome models and poor performance at the segment assemblage stage. To address these problems, we propose M-SAN (Multi-modal Segment Assemblage Network) which can perform efficient and coherent segment assemblage task end-to-end. It utilizes multi-modal representation extracted from the segments and follows the Encoder-Decoder Ptr-Net framework with the Attention mechanism. Importance-coherence reward is designed for training M-SAN. We experiment on the Ads-1k dataset with 1000+ videos under rich ad scenarios collected from advertisers. To evaluate the methods, we propose a unified metric, Imp-Coh@Time, which comprehensively assesses the importance, coherence, and duration of the outputs at the same time. Experimental results show that our method achieves better performance than random selection and the previous method on the metric. Ablation experiments further verify that multi-modal representation and importance-coherence reward significantly improve the performance. Ads-1k dataset is available at: https://github.com/yunlong10/Ads-1k

arXiv:2209.12043

Unsupervised domain adaptation for speech recognition with unsupervised error correction

The transcription quality of automatic speech recognition (ASR) systems degrades significantly when transcribing audios coming from unseen domains. We propose an unsupervised error correction method for unsupervised ASR domain adaption, aiming to recover transcription errors caused by domain mismatch. Unlike existing correction methods that rely on transcribed audios for training, our approach requires only unlabeled data of the target domains in which a pseudo-labeling technique is applied to generate correction training samples. To reduce over-fitting to the pseudo data, we also propose an encoder-decoder correction model that can take into account additional information such as dialogue context and acoustic features. Experiment results show that our method obtains a significant word error rate (WER) reduction over non-adapted ASR systems. The correction model can also be applied on top of other adaptation approaches to bring an additional improvement of 10% relatively.

arXiv:2209.11395

Achieve the Minimum Width of Neural Networks for Universal Approximation

The universal approximation property (UAP) of neural networks is fundamental for deep learning, and it is well known that wide neural networks are universal approximators of continuous functions within both the $L^p$ norm and the continuous/uniform norm. However, the exact minimum width, $w_{\min}$, for the UAP has not been studied thoroughly. Recently, using a decoder-memorizer-encoder scheme, \citet{Park2021Minimum} found that $w_{\min} = \max(d_x+1,d_y)$ for both the $L^p$-UAP of ReLU networks and the $C$-UAP of ReLU+STEP networks, where $d_x,d_y$ are the input and output dimensions, respectively. In this paper, we consider neural networks with an arbitrary set of activation functions. We prove that both $C$-UAP and $L^p$-UAP for functions on compact domains share a universal lower bound of the minimal width; that is, $w^*_{\min} = \max(d_x,d_y)$. In particular, the critical width, $w^*_{\min}$, for $L^p$-UAP can be achieved by leaky-ReLU networks, provided that the input or output dimension is larger than one. Our construction is based on the approximation power of neural ordinary differential equations and the ability to approximate flow maps by neural networks. The nonmonotone or discontinuous activation functions case and the one-dimensional case are also discussed.

arXiv:2209.11255

3DGTN: 3D Dual-Attention GLocal Transformer Network for Point Cloud Classification and Segmentation

Although the application of Transformers in 3D point cloud processing has achieved significant progress and success, it is still challenging for existing 3D Transformer methods to efficiently and accurately learn both valuable global features and valuable local features for improved applications. This paper presents a novel point cloud representational learning network, called 3D Dual Self-attention Global Local (GLocal) Transformer Network (3DGTN), for improved feature learning in both classification and segmentation tasks, with the following key contributions. First, a GLocal Feature Learning (GFL) block with the dual self-attention mechanism (i.e., a novel Point-Patch Self-Attention, called PPSA, and a channel-wise self-attention) is designed to efficiently learn the GLocal context information. Second, the GFL block is integrated with a multi-scale Graph Convolution-based Local Feature Aggregation (LFA) block, leading to a Global-Local (GLocal) information extraction module that can efficiently capture critical information. Third, a series of GLocal modules are used to construct a new hierarchical encoder-decoder structure to enable the learning of "GLocal" information in different scales in a hierarchical manner. The proposed framework is evaluated on both classification and segmentation datasets, demonstrating that the proposed method is capable of outperforming many state-of-the-art methods on both classification and segmentation tasks.

arXiv:2209.10712공식 코드 · Papers with Code

Compressing Sign Information in DCT-based Image Coding via Deep Sign Retrieval

Compressing the sign information of discrete cosine transform (DCT) coefficients is an intractable problem in image coding schemes due to the equiprobable characteristics of the signs. To overcome this difficulty, we propose an efficient compression method for the sign information called "sign retrieval." This method is inspired by phase retrieval, which is a classical signal restoration problem of finding the phase information of discrete Fourier transform coefficients from their magnitudes. The sign information of all DCT coefficients is excluded from a bitstream at the encoder and is complemented at the decoder through our sign retrieval method. We show through experiments that our method outperforms previous ones in terms of the bit amount for the signs and computation cost. Our method, implemented in Python language, is available from https://github.com/ctsutake/dsr.

arXiv:2209.10922

Learning to Write with Coherence From Negative Examples

Coherence is one of the critical factors that determine the quality of writing. We propose writing relevance (WR) training method for neural encoder-decoder natural language generation (NLG) models which improves coherence of the continuation by leveraging negative examples. WR loss regresses the vector representation of the context and generated sentence toward positive continuation by contrasting it with the negatives. We compare our approach with Unlikelihood (UL) training in a text continuation task on commonsense natural language inference (NLI) corpora to show which method better models the coherence by avoiding unlikely continuations. The preference of our approach in human evaluation shows the efficacy of our method in improving coherence.

arXiv:2209.10963

COVID-19 Detection and Analysis From Lung CT Images using Novel Channel Boosted CNNs

In December 2019, the global pandemic COVID-19 in Wuhan, China, affected human life and the worldwide economy. Therefore, an efficient diagnostic system is required to control its spread. However, the automatic diagnostic system poses challenges with a limited amount of labeled data, minor contrast variation, and high structural similarity between infection and background. In this regard, a new two-phase deep convolutional neural network (CNN) based diagnostic system is proposed to detect minute irregularities and analyze COVID-19 infection. In the first phase, a novel SB-STM-BRNet CNN is developed, incorporating a new channel Squeezed and Boosted (SB) and dilated convolutional-based Split-Transform-Merge (STM) block to detect COVID-19 infected lung CT images. The new STM blocks performed multi-path region-smoothing and boundary operations, which helped to learn minor contrast variation and global COVID-19 specific patterns. Furthermore, the diverse boosted channels are achieved using the SB and Transfer Learning concepts in STM blocks to learn texture variation between COVID-19-specific and healthy images. In the second phase, COVID-19 infected images are provided to the novel COVID-CB-RESeg segmentation CNN to identify and analyze COVID-19 infectious regions. The proposed COVID-CB-RESeg methodically employed region-homogeneity and heterogeneity operations in each encoder-decoder block and boosted-decoder using auxiliary channels to simultaneously learn the low illumination and boundaries of the COVID-19 infected region. The proposed diagnostic system yields good performance in terms of accuracy: 98.21 %, F-score: 98.24%, Dice Similarity: 96.40 %, and IOU: 98.85 % for the COVID-19 infected region. The proposed diagnostic system would reduce the burden and strengthen the radiologist's decision for a fast and accurate COVID-19 diagnosis.