PyTorchKR Papers

최신 논문

총 9,339편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2211.09676공식 코드 · Papers with Code

Verified Reversible Programming for Verified Lossless Compression

Lossless compression implementations typically contain two programs, an encoder and a decoder, which are required to be inverse to one another. We observe that a significant class of compression methods, based on asymmetric numeral systems (ANS), have shared structure between the encoder and decoder -- the decoder program is the 'reverse' of the encoder program -- allowing both to be simultaneously specified by a single, reversible function. To exploit this, we have implemented a small reversible language, embedded in Agda, which we call 'Flipper' (available at https://github.com/j-towns/flipper). Agda supports formal verification of program properties, and the compiler for our reversible language (which is implemented as an Agda macro), produces not just an encoder/decoder pair of functions but also a proof that they are inverse to one another. Thus users of the language get formal verification 'for free'. We give a small example use-case of Flipper in this paper, and plan to publish a full compression implementation soon.

arXiv:2211.09458공식 코드 · Papers with Code

Abstractive Summarization Guided by Latent Hierarchical Document Structure

Sequential abstractive neural summarizers often do not use the underlying structure in the input article or dependencies between the input sentences. This structure is essential to integrate and consolidate information from different parts of the text. To address this shortcoming, we propose a hierarchy-aware graph neural network (HierGNN) which captures such dependencies through three main steps: 1) learning a hierarchical document structure through a latent structure tree learned by a sparse matrix-tree computation; 2) propagating sentence information over this structure using a novel message-passing node propagation mechanism to identify salient information; 3) using graph-level attention to concentrate the decoder on salient information. Experiments confirm HierGNN improves strong sequence models such as BART, with a 0.55 and 0.75 margin in average ROUGE-1/2/L for CNN/DM and XSum. Further human evaluation demonstrates that summaries produced by our model are more relevant and less redundant than the baselines, into which HierGNN is incorporated. We also find HierGNN synthesizes summaries by fusing multiple source sentences more, rather than compressing a single source sentence, and that it processes long inputs more effectively.

arXiv:2211.09808공식 코드 · Papers with Code

Uni-Perceiver v2: A Generalist Model for Large-Scale Vision and Vision-Language Tasks

Despite the remarkable success of foundation models, their task-specific fine-tuning paradigm makes them inconsistent with the goal of general perception modeling. The key to eliminating this inconsistency is to use generalist models for general task modeling. However, existing attempts at generalist models are inadequate in both versatility and performance. In this paper, we propose Uni-Perceiver v2, which is the first generalist model capable of handling major large-scale vision and vision-language tasks with competitive performance. Specifically, images are encoded as general region proposals, while texts are encoded via a Transformer-based language model. The encoded representations are transformed by a task-agnostic decoder. Different tasks are formulated as a unified maximum likelihood estimation problem. We further propose an improved optimizer to ensure stable multi-task learning with an unmixed sampling strategy, which is helpful for tasks requiring large batch-size training. After being jointly trained on various tasks, Uni-Perceiver v2 is capable of directly handling downstream tasks without any task-specific adaptation. Results show that Uni-Perceiver v2 outperforms all existing generalist models in both versatility and performance. Meanwhile, compared with the commonly-recognized strong baselines that require tasks-specific fine-tuning, Uni-Perceiver v2 achieves competitive performance on a broad range of vision and vision-language tasks.

arXiv:2211.09768

D$^3$ETR: Decoder Distillation for Detection Transformer

While various knowledge distillation (KD) methods in CNN-based detectors show their effectiveness in improving small students, the baselines and recipes for DETR-based detectors are yet to be built. In this paper, we focus on the transformer decoder of DETR-based detectors and explore KD methods for them. The outputs of the transformer decoder lie in random order, which gives no direct correspondence between the predictions of the teacher and the student, thus posing a challenge for knowledge distillation. To this end, we propose MixMatcher to align the decoder outputs of DETR-based teachers and students, which mixes two teacher-student matching strategies, i.e., Adaptive Matching and Fixed Matching. Specifically, Adaptive Matching applies bipartite matching to adaptively match the outputs of the teacher and the student in each decoder layer, while Fixed Matching fixes the correspondence between the outputs of the teacher and the student with the same object queries, with the teacher's fixed object queries fed to the decoder of the student as an auxiliary group. Based on MixMatcher, we build \textbf{D}ecoder \textbf{D}istillation for \textbf{DE}tection \textbf{TR}ansformer (D$^3$ETR), which distills knowledge in decoder predictions and attention maps from the teachers to students. D$^3$ETR shows superior performance on various DETR-based detectors with different backbones. For example, D$^3$ETR improves Conditional DETR-R50-C5 by $\textbf{7.8}/\textbf{2.4}$ mAP under $12/50$ epochs training settings with Conditional DETR-R101-C5 as the teacher.

arXiv:2211.09159공식 코드 · Papers with Code

Unified Question Answering in Slovene

Question answering is one of the most challenging tasks in language understanding. Most approaches are developed for English, while less-resourced languages are much less researched. We adapt a successful English question-answering approach, called UnifiedQA, to the less-resourced Slovene language. Our adaptation uses the encoder-decoder transformer SloT5 and mT5 models to handle four question-answering formats: yes/no, multiple-choice, abstractive, and extractive. We use existing Slovene adaptations of four datasets, and machine translate the MCTest dataset. We show that a general model can answer questions in different formats at least as well as specialized models. The results are further improved using cross-lingual transfer from English. While we produce state-of-the-art results for Slovene, the performance still lags behind English.

arXiv:2211.08726공식 코드 · Papers with Code

Streaming Joint Speech Recognition and Disfluency Detection

Disfluency detection has mainly been solved in a pipeline approach, as post-processing of speech recognition. In this study, we propose Transformer-based encoder-decoder models that jointly solve speech recognition and disfluency detection, which work in a streaming manner. Compared to pipeline approaches, the joint models can leverage acoustic information that makes disfluency detection robust to recognition errors and provide non-verbal clues. Moreover, joint modeling results in low-latency and lightweight inference. We investigate two joint model variants for streaming disfluency detection: a transcript-enriched model and a multi-task model. The transcript-enriched model is trained on text with special tags indicating the starting and ending points of the disfluent part. However, it has problems with latency and standard language model adaptation, which arise from the additional disfluency tags. We propose a multi-task model to solve such problems, which has two output layers at the Transformer decoder; one for speech recognition and the other for disfluency detection. It is modeled to be conditioned on the currently recognized token with an additional token-dependency mechanism. We show that the proposed joint models outperformed a BERT-based pipeline approach in both accuracy and latency, on both the Switchboard and the corpus of spontaneous Japanese.

arXiv:2211.08634공식 코드 · Papers with Code

Semantic keypoint extraction for scanned animals using multi-depth-camera systems

Keypoint annotation in point clouds is an important task for 3D reconstruction, object tracking and alignment, in particular in deformable or moving scenes. In the context of agriculture robotics, it is a critical task for livestock automation to work toward condition assessment or behaviour recognition. In this work, we propose a novel approach for semantic keypoint annotation in point clouds, by reformulating the keypoint extraction as a regression problem of the distance between the keypoints and the rest of the point cloud. We use the distance on the point cloud manifold mapped into a radial basis function (RBF), which is then learned using an encoder-decoder architecture. Special consideration is given to the data augmentation specific to multi-depth-camera systems by considering noise over the extrinsic calibration and camera frame dropout. Additionally, we investigate computationally efficient non-rigid deformation methods that can be applied to animal point clouds. Our method is tested on data collected in the field, on moving beef cattle, with a calibrated system of multiple hardware-synchronised RGB-D cameras.

arXiv:2211.08428

CaDM: Codec-aware Diffusion Modeling for Neural-enhanced Video Streaming

Recent years have witnessed the dramatic growth of Internet video traffic, where the video bitstreams are often compressed and delivered in low quality to fit the streamer's uplink bandwidth. To alleviate the quality degradation, it comes the rise of Neural-enhanced Video Streaming (NVS), which shows great prospects for recovering low-quality videos by mostly deploying neural super-resolution (SR) on the media server. Despite its benefit, we reveal that current mainstream works with SR enhancement have not achieved the desired rate-distortion trade-off between bitrate saving and quality restoration, due to: (1) overemphasizing the enhancement on the decoder side while omitting the co-design of encoder, (2) limited generative capacity to recover high-fidelity perceptual details, and (3) optimizing the compression-and-restoration pipeline from the resolution perspective solely, without considering color bit-depth. Aiming at overcoming these limitations, we are the first to conduct an encoder-decoder (i.e., codec) synergy by leveraging the inherent visual-generative property of diffusion models. Specifically, we present the Codec-aware Diffusion Modeling (CaDM), a novel NVS paradigm to significantly reduce streaming delivery bitrates while holding pretty higher restoration capacity over existing methods. First, CaDM improves the encoder's compression efficiency by simultaneously reducing resolution and color bit-depth of video frames. Second, CaDM empowers the decoder with high-quality enhancement by making the denoising diffusion restoration aware of encoder's resolution-color conditions. Evaluation on public cloud services with OpenMMLab benchmarks shows that CaDM effectively saves up to 5.12 - 21.44 times bitrates based on common video standards and achieves much better recovery quality (e.g., FID of 0.61) over state-of-the-art neural-enhancing methods.

arXiv:2211.08843

Improving Speech Emotion Recognition with Unsupervised Speaking Style Transfer

Humans can effortlessly modify various prosodic attributes, such as the placement of stress and the intensity of sentiment, to convey a specific emotion while maintaining consistent linguistic content. Motivated by this capability, we propose EmoAug, a novel style transfer model designed to enhance emotional expression and tackle the data scarcity issue in speech emotion recognition tasks. EmoAug consists of a semantic encoder and a paralinguistic encoder that represent verbal and non-verbal information respectively. Additionally, a decoder reconstructs speech signals by conditioning on the aforementioned two information flows in an unsupervised fashion. Once training is completed, EmoAug enriches expressions of emotional speech with different prosodic attributes, such as stress, rhythm and intensity, by feeding different styles into the paralinguistic encoder. EmoAug enables us to generate similar numbers of samples for each class to tackle the data imbalance issue as well. Experimental results on the IEMOCAP dataset demonstrate that EmoAug can successfully transfer different speaking styles while retaining the speaker identity and semantic content. Furthermore, we train a SER model with data augmented by EmoAug and show that the augmented model not only surpasses the state-of-the-art supervised and self-supervised methods but also overcomes overfitting problems caused by data imbalance. Some audio samples can be found on our demo website.

arXiv:2211.08717

SWIN-SFTNet : Spatial Feature Expansion and Aggregation using Swin Transformer For Whole Breast micro-mass segmentation

Incorporating various mass shapes and sizes in training deep learning architectures has made breast mass segmentation challenging. Moreover, manual segmentation of masses of irregular shapes is time-consuming and error-prone. Though Deep Neural Network has shown outstanding performance in breast mass segmentation, it fails in segmenting micro-masses. In this paper, we propose a novel U-net-shaped transformer-based architecture, called Swin-SFTNet, that outperforms state-of-the-art architectures in breast mammography-based micro-mass segmentation. Firstly to capture the global context, we designed a novel Spatial Feature Expansion and Aggregation Block(SFEA) that transforms sequential linear patches into a structured spatial feature. Next, we combine it with the local linear features extracted by the swin transformer block to improve overall accuracy. We also incorporate a novel embedding loss that calculates similarities between linear feature embeddings of the encoder and decoder blocks. With this approach, we achieve higher segmentation dice over the state-of-the-art by 3.10% on CBIS-DDSM, 3.81% on InBreast, and 3.13% on CBIS pre-trained model on the InBreast test data set.

arXiv:2211.07743공식 코드 · Papers with Code

Generative Aspect-Based Sentiment Analysis with Contrastive Learning and Expressive Structure

Generative models have demonstrated impressive results on Aspect-based Sentiment Analysis (ABSA) tasks, particularly for the emerging task of extracting Aspect-Category-Opinion-Sentiment (ACOS) quadruples. However, these models struggle with implicit sentiment expressions, which are commonly observed in opinionated content such as online reviews. In this work, we introduce GEN-SCL-NAT, which consists of two techniques for improved structured generation for ACOS quadruple extraction. First, we propose GEN-SCL, a supervised contrastive learning objective that aids quadruple prediction by encouraging the model to produce input representations that are discriminable across key input attributes, such as sentiment polarity and the existence of implicit opinions and aspects. Second, we introduce GEN-NAT, a new structured generation format that better adapts autoregressive encoder-decoder models to extract quadruples in a generative fashion. Experimental results show that GEN-SCL-NAT achieves top performance across three ACOS datasets, averaging 1.48% F1 improvement, with a maximum 1.73% increase on the LAPTOP-L1 dataset. Additionally, we see significant gains on implicit aspect and opinion splits that have been shown as challenging for existing ACOS approaches.

arXiv:2211.07703

Alzheimer's Diagnosis and Generation-Based Chatbot Using Hierarchical Attention and Transformer

In this paper, we propose a natural language processing architecture that can handle tasks that previously required two models as one model. With a single model, we analyze the language patterns and conversational context of Alzheimer's patients and derive answers from two results: patient classification and chatbot. If the patient's language characteristics are identified by chatbots in daily life, doctors can plan more precise diagnosis and treatment for early diagnosis. The proposed model is used to develop chatbots that replace questionnaires that required experts. There are two natural language processing tasks performed by the model. The first is a 'natural language classification' that indicates with probability whether the patient has an illness, and the second is to generate the next 'answer' of the chatbot to the patient's answer. In the first half, a context vector, which is a characteristic of patient utterance, is extracted through a self-attention neural network. This context vector and chatbot (expert, moderator) questions are entered together into the encoder to obtain a matrix containing the characteristics of the interaction between the questioner and the patient. The vectorized matrix becomes a probability value for classification of patients. Enter the matrix into the decoder with the next answer from the chatbot (the moderator) to generate the next utterance. As a result of learning this structure with DmentiaBank's cookie theft description corpus, it was confirmed that the value of the loss function of the encoder and decoder was significantly reduced and converged. This shows that capturing the speech language pattern of Alzheimer's disease patients can contribute to early diagnosis and longitudinal studies of the disease in the future.

arXiv:2211.07891

Feedback Chain Network For Hippocampus Segmentation

The hippocampus plays a vital role in the diagnosis and treatment of many neurological disorders. Recent years, deep learning technology has made great progress in the field of medical image segmentation, and the performance of related tasks has been constantly refreshed. In this paper, we focus on the hippocampus segmentation task and propose a novel hierarchical feedback chain network. The feedback chain structure unit learns deeper and wider feature representation of each encoder layer through the hierarchical feature aggregation feedback chains, and achieves feature selection and feedback through the feature handover attention module. Then, we embed a global pyramid attention unit between the feature encoder and the decoder to further modify the encoder features, including the pair-wise pyramid attention module for achieving adjacent attention interaction and the global context modeling module for capturing the long-range knowledge. The proposed approach achieves state-of-the-art performance on three publicly available datasets, compared with existing hippocampus segmentation approaches.

arXiv:2211.08115공식 코드 · Papers with Code

Heatmap-based Out-of-Distribution Detection

Our work investigates out-of-distribution (OOD) detection as a neural network output explanation problem. We learn a heatmap representation for detecting OOD images while visualizing in- and out-of-distribution image regions at the same time. Given a trained and fixed classifier, we train a decoder neural network to produce heatmaps with zero response for in-distribution samples and high response heatmaps for OOD samples, based on the classifier features and the class prediction. Our main innovation lies in the heatmap definition for an OOD sample, as the normalized difference from the closest in-distribution sample. The heatmap serves as a margin to distinguish between in- and out-of-distribution samples. Our approach generates the heatmaps not only for OOD detection, but also to indicate in- and out-of-distribution regions of the input image. In our evaluations, our approach mostly outperforms the prior work on fixed classifiers, trained on CIFAR-10, CIFAR-100 and Tiny ImageNet. The code is publicly available at: https://github.com/jhornauer/heatmap_ood.

arXiv:2211.07968공식 코드 · Papers with Code

NeRFFaceEditing: Disentangled Face Editing in Neural Radiance Fields

Recent methods for synthesizing 3D-aware face images have achieved rapid development thanks to neural radiance fields, allowing for high quality and fast inference speed. However, existing solutions for editing facial geometry and appearance independently usually require retraining and are not optimized for the recent work of generation, thus tending to lag behind the generation process. To address these issues, we introduce NeRFFaceEditing, which enables editing and decoupling geometry and appearance in the pretrained tri-plane-based neural radiance field while retaining its high quality and fast inference speed. Our key idea for disentanglement is to use the statistics of the tri-plane to represent the high-level appearance of its corresponding facial volume. Moreover, we leverage a generated 3D-continuous semantic mask as an intermediary for geometry editing. We devise a geometry decoder (whose output is unchanged when the appearance changes) and an appearance decoder. The geometry decoder aligns the original facial volume with the semantic mask volume. We also enhance the disentanglement by explicitly regularizing rendered images with the same appearance but different geometry to be similar in terms of color distribution for each facial component separately. Our method allows users to edit via semantic masks with decoupled control of geometry and appearance. Both qualitative and quantitative evaluations show the superior geometry and appearance control abilities of our method compared to existing and alternative solutions.

arXiv:2211.08279공식 코드 · Papers with Code

Towards an objective characterization of an individual's facial movements using Self-Supervised Person-Specific-Models

Disentangling facial movements from other facial characteristics, particularly from facial identity, remains a challenging task, as facial movements display great variation between individuals. In this paper, we aim to characterize individual-specific facial movements. We present a novel training approach to learn facial movements independently of other facial characteristics, focusing on each individual separately. We propose self-supervised Person-Specific Models (PSMs), in which one model per individual can learn to extract an embedding of the facial movements independently of the person's identity and other structural facial characteristics from unlabeled facial video. These models are trained using encoder-decoder-like architectures. We provide quantitative and qualitative evidence that a PSM learns a meaningful facial embedding that discovers fine-grained movements otherwise not characterized by a General Model (GM), which is trained across individuals and characterizes general patterns of facial movements. We present quantitative and qualitative evidence that this approach is easily scalable and generalizable for new individuals: facial movements knowledge learned on a person can quickly and effectively be transferred to a new person. Lastly, we propose a novel PSM using curriculum temporal learning to leverage the temporal contiguity between video frames. Our code, analysis details, and all pretrained models are available in Github and Supplementary Materials.

arXiv:2211.07082

Learning Latent Part-Whole Hierarchies for Point Clouds

Strong evidence suggests that humans perceive the 3D world by parsing visual scenes and objects into part-whole hierarchies. Although deep neural networks have the capability of learning powerful multi-level representations, they can not explicitly model part-whole hierarchies, which limits their expressiveness and interpretability in processing 3D vision data such as point clouds. To this end, we propose an encoder-decoder style latent variable model that explicitly learns the part-whole hierarchies for the multi-level point cloud segmentation. Specifically, the encoder takes a point cloud as input and predicts the per-point latent subpart distribution at the middle level. The decoder takes the latent variable and the feature from the encoder as an input and predicts the per-point part distribution at the top level. During training, only annotated part labels at the top level are provided, thus making the whole framework weakly supervised. We explore two kinds of approximated inference algorithms, i.e., most-probable-latent and Monte Carlo methods, and three stochastic gradient estimations for learning discrete latent variables, i.e., straight-through, REINFORCE, and pathwise estimators. Experimental results on the PartNet dataset show that the proposed method achieves state-of-the-art performance in not only top-level part segmentation but also middle-level latent subpart segmentation.

arXiv:2211.06841공식 코드 · Papers with Code

Point-DAE: Denoising Autoencoders for Self-supervised Point Cloud Learning

Masked autoencoder has demonstrated its effectiveness in self-supervised point cloud learning. Considering that masking is a kind of corruption, in this work we explore a more general denoising autoencoder for point cloud learning (Point-DAE) by investigating more types of corruptions beyond masking. Specifically, we degrade the point cloud with certain corruptions as input, and learn an encoder-decoder model to reconstruct the original point cloud from its corrupted version. Three corruption families (\ie, density/masking, noise, and affine transformation) and a total of fourteen corruption types are investigated with traditional non-Transformer encoders. Besides the popular masking corruption, we identify another effective corruption family, \ie, affine transformation. The affine transformation disturbs all points globally, which is complementary to the masking corruption where some local regions are dropped. We also validate the effectiveness of affine transformation corruption with the Transformer backbones, where we decompose the reconstruction of the complete point cloud into the reconstructions of detailed local patches and rough global shape, alleviating the position leakage problem in the reconstruction. Extensive experiments on tasks of object classification, few-shot learning, robustness testing, part segmentation, and 3D object detection validate the effectiveness of the proposed method. The codes are available at \url{https://github.com/YBZh/Point-DAE}.

arXiv:2211.07210

Grafting Pre-trained Models for Multimodal Headline Generation

Multimodal headline utilizes both video frames and transcripts to generate the natural language title of the videos. Due to a lack of large-scale, manually annotated data, the task of annotating grounded headlines for video is labor intensive and impractical. Previous researches on pre-trained language models and video-language models have achieved significant progress in related downstream tasks. However, none of them can be directly applied to multimodal headline architecture where we need both multimodal encoder and sentence decoder. A major challenge in simply gluing language model and video-language model is the modality balance, which is aimed at combining visual-language complementary abilities. In this paper, we propose a novel approach to graft the video encoder from the pre-trained video-language model on the generative pre-trained language model. We also present a consensus fusion mechanism for the integration of different components, via inter/intra modality relation. Empirically, experiments show that the grafted model achieves strong results on a brand-new dataset collected from real-world applications.

arXiv:2211.06962

A Scalable Graph Neural Network Decoder for Short Block Codes

In this work, we propose a novel decoding algorithm for short block codes based on an edge-weighted graph neural network (EW-GNN). The EW-GNN decoder operates on the Tanner graph with an iterative message-passing structure, which algorithmically aligns with the conventional belief propagation (BP) decoding method. In each iteration, the "weight" on the message passed along each edge is obtained from a fully connected neural network that has the reliability information from nodes/edges as its input. Compared to existing deep-learning-based decoding schemes, the EW-GNN decoder is characterised by its scalability, meaning that 1) the number of trainable parameters is independent of the codeword length, and 2) an EW-GNN decoder trained with shorter/simple codes can be directly used for longer/sophisticated codes of different code rates. Furthermore, simulation results show that the EW-GNN decoder outperforms the BP and deep-learning-based BP methods from the literature in terms of the decoding error rate.