Recent works on knowledge base question answering (KBQA) retrieve subgraphs for easier reasoning. A desired subgraph is crucial as a small one may exclude the answer but a large one might introduce more noises. However, the existing retrieval is either heuristic or interwoven with the reasoning, causing reasoning on the partial subgraphs, which increases the reasoning bias when the intermediate supervision is missing. This paper proposes a trainable subgraph retriever (SR) decoupled from the subsequent reasoning process, which enables a plug-and-play framework to enhance any subgraph-oriented KBQA model. Extensive experiments demonstrate SR achieves significantly better retrieval and QA performance than existing retrieval methods. Via weakly supervised pre-training as well as the end-to-end fine-tuning, SRl achieves new state-of-the-art performance when combined with NSM, a subgraph-oriented reasoner, for embedding-based KBQA methods.
Vision transformers have shown great potential in various computer vision tasks owing to their strong capability to model long-range dependency using the self-attention mechanism. Nevertheless, they treat an image as a 1D sequence of visual tokens, lacking an intrinsic inductive bias (IB) in modeling local visual structures and dealing with scale variance, which is instead learned implicitly from large-scale training data with longer training schedules. In this paper, we propose a Vision Transformer Advanced by Exploring intrinsic IB from convolutions, i.e., ViTAE. Technically, ViTAE has several spatial pyramid reduction modules to downsample and embed the input image into tokens with rich multi-scale context using multiple convolutions with different dilation rates. In this way, it acquires an intrinsic scale invariance IB and can learn robust feature representation for objects at various scales. Moreover, in each transformer layer, ViTAE has a convolution block parallel to the multi-head self-attention module, whose features are fused and fed into the feed-forward network. Consequently, it has the intrinsic locality IB and is able to learn local features and global dependencies collaboratively. The proposed two kinds of cells are stacked in both isotropic and multi-stage manners to formulate two families of ViTAE models, i.e., the vanilla ViTAE and ViTAEv2. Experiments on the ImageNet dataset as well as downstream tasks on the MS COCO, ADE20K, and AP10K datasets validate the superiority of our models over the baseline transformer models and concurrent works. Besides, we scale up our ViTAE model to 644M parameters and obtain the state-of-the-art classification performance, i.e., 88.5% Top-1 classification accuracy on ImageNet validation set and the best 91.2% Top-1 accuracy on ImageNet real validation set, without using extra private data.
Although point-based networks are demonstrated to be accurate for 3D point cloud modeling, they are still falling behind their voxel-based competitors in 3D detection. We observe that the prevailing set abstraction design for down-sampling points may maintain too much unimportant background information that can affect feature learning for detecting objects. To tackle this issue, we propose a novel set abstraction method named Semantics-Augmented Set Abstraction (SASA). Technically, we first add a binary segmentation module as the side output to help identify foreground points. Based on the estimated point-wise foreground scores, we then propose a semantics-guided point sampling algorithm to help retain more important foreground points during down-sampling. In practice, SASA shows to be effective in identifying valuable points related to foreground objects and improving feature learning for point-based 3D detection. Additionally, it is an easy-to-plug-in module and able to boost various point-based detectors, including single-stage and two-stage ones. Extensive experiments on the popular KITTI and nuScenes datasets validate the superiority of SASA, lifting point-based detection models to reach comparable performance to state-of-the-art voxel-based methods.
The exemplar-free class incremental learning requires classification models to learn new class knowledge incrementally without retaining any old samples. Recently, the framework based on parallel one-class classifiers (POC), which trains a one-class classifier (OCC) independently for each category, has attracted extensive attention, since it can naturally avoid catastrophic forgetting. POC, however, suffers from weak discriminability and comparability due to its independent training strategy for different OOCs. To meet this challenge, we propose a new framework, named Discriminative and Comparable One-class classifiers for Incremental Learning (DisCOIL). DisCOIL follows the basic principle of POC, but it adopts variational auto-encoders (VAE) instead of other well-established one-class classifiers (e.g. deep SVDD), because a trained VAE can not only identify the probability of an input sample belonging to a class but also generate pseudo samples of the class to assist in learning new tasks. With this advantage, DisCOIL trains a new-class VAE in contrast with the old-class VAEs, which forces the new-class VAE to reconstruct better for new-class samples but worse for the old-class pseudo samples, thus enhancing the comparability. Furthermore, DisCOIL introduces a hinge reconstruction loss to ensure the discriminability. We evaluate our method extensively on MNIST, CIFAR10, and Tiny-ImageNet. The experimental results show that DisCOIL achieves state-of-the-art performance.
The success of existing salient object detection models relies on a large pixel-wise labeled training dataset. How-ever, collecting such a dataset is not only time-consuming but also very expensive. To reduce the labeling burden, we study semi-supervised salient object detection, and formulate it as an unlabeled dataset pixel-level confidence estimation problem by identifying pixels with less confident predictions. Specifically, we introduce a new latent variable model with an energy-based prior for effective latent space exploration, leading to more reliable confidence maps. With the proposed strategy, the unlabelled images can effectively participate in model training. Experimental results show that the proposed solution, using only 1/16 of the annotations from the original training dataset, achieves competitive performance compared with state-of-the-art fully supervised models.
Semi-supervised video object segmentation (VOS) refers to segmenting the target object in remaining frames given its annotation in the first frame, which has been actively studied in recent years. The key challenge lies in finding effective ways to exploit the spatio-temporal context of past frames to help learn discriminative target representation of current frame. In this paper, we propose a novel Siamese network with a specifically designed interactive transformer, called SITVOS, to enable effective context propagation from historical to current frames. Technically, we use the transformer encoder and decoder to handle the past frames and current frame separately, i.e., the encoder encodes robust spatio-temporal context of target object from the past frames, while the decoder takes the feature embedding of current frame as the query to retrieve the target from the encoder output. To further enhance the target representation, a feature interaction module (FIM) is devised to promote the information flow between the encoder and decoder. Moreover, we employ the Siamese architecture to extract backbone features of both past and current frames, which enables feature reuse and is more efficient than existing methods. Experimental results on three challenging benchmarks validate the superiority of SITVOS over state-of-the-art methods.
Different from large-scale platforms such as Taobao and Amazon, developing CVR models in small-scale recommendation scenarios is more challenging due to the severe Data Distribution Fluctuation (DDF) issue. DDF prevents existing CVR models from being effective since 1) several months of data are needed to train CVR models sufficiently in small scenarios, leading to considerable distribution discrepancy between training and online serving; and 2) e-commerce promotions have much more significant impacts on small scenarios, leading to distribution uncertainty of the upcoming time period. In this work, we propose a novel CVR method named MetaCVR from a perspective of meta learning to address the DDF issue. Firstly, a base CVR model which consists of a Feature Representation Network (FRN) and output layers is elaborately designed and trained sufficiently with samples across months. Then we treat time periods with different data distributions as different occasions and obtain positive and negative prototypes for each occasion using the corresponding samples and the pre-trained FRN. Subsequently, a Distance Metric Network (DMN) is devised to calculate the distance metrics between each sample and all prototypes to facilitate mitigating the distribution uncertainty. At last, we develop an Ensemble Prediction Network (EPN) which incorporates the output of FRN and DMN to make the final CVR prediction. In this stage, we freeze the FRN and train the DMN and EPN with samples from recent time period, therefore effectively easing the distribution discrepancy. To the best of our knowledge, this is the first study of CVR prediction targeting the DDF issue in small-scale recommendation scenarios. Experimental results on real-world datasets validate the superiority of our MetaCVR and online A/B test also shows our model achieves impressive gains of 11.92% on PCVR and 8.64% on GMV.
Promotions are becoming more important and prevalent in e-commerce platforms to attract customers and boost sales. However, Click-Through Rate (CTR) prediction methods in recommender systems are not able to handle such circumstances well since: 1) they can't generalize well to serving because the online data distribution is uncertain due to the potentially upcoming promotions; 2) without paying enough attention to scenario signals, they are incapable of learning different feature representation patterns which coexist in each scenario. In this work, we propose Scenario Adaptive Mixture-of-Experts (SAME), a simple yet effective model that serves both promotion and normal scenarios. Technically, it follows the idea of Mixture-of-Experts by adopting multiple experts to learn feature representations, which are modulated by a Feature Gated Network (FGN) via an attention mechanism. To obtain high-quality representations, we design a Stacked Parallel Attention Unit (SPAU) to help each expert better handle user behavior sequence. To tackle the distribution uncertainty, a set of scenario signals are elaborately devised from a perspective of time series prediction and fed into the FGN, whose output is concatenated with feature representation from each expert to learn the attention. Accordingly, a mixture of the feature representations is obtained scenario-adaptively and used for the final CTR prediction. In this way, each expert can learn a discriminative representation pattern. To the best of our knowledge, this is the first study for promotion-aware CTR prediction. Experimental results on real-world datasets validate the superiority of SAME. Online A/B test also shows SAME achieves significant gains of 3.58% on CTR and 5.94% on IPV during promotion periods as well as 3.93% and 6.57% in normal days, respectively.
Vision transformer networks have shown superiority in many computer vision tasks. In this paper, we take a step further by proposing a novel generative vision transformer with latent variables following an informative energy-based prior for salient object detection. Both the vision transformer network and the energy-based prior model are jointly trained via Markov chain Monte Carlo-based maximum likelihood estimation, in which the sampling from the intractable posterior and prior distributions of the latent variables are performed by Langevin dynamics. Further, with the generative vision transformer, we can easily obtain a pixel-wise uncertainty map from an image, which indicates the model confidence in predicting saliency from the image. Different from the existing generative models which define the prior distribution of the latent variables as a simple isotropic Gaussian distribution, our model uses an energy-based informative prior which can be more expressive to capture the latent space of the data. We apply the proposed framework to both RGB and RGB-D salient object detection tasks. Extensive experimental results show that our framework can achieve not only accurate saliency predictions but also meaningful uncertainty maps that are consistent with the human perception.
Existing Scene Text Recognition (STR) methods typically use a language model to optimize the joint probability of the 1D character sequence predicted by a visual recognition (VR) model, which ignore the 2D spatial context of visual semantics within and between character instances, making them not generalize well to arbitrary shape scene text. To address this issue, we make the first attempt to perform textual reasoning based on visual semantics in this paper. Technically, given the character segmentation maps predicted by a VR model, we construct a subgraph for each instance, where nodes represent the pixels in it and edges are added between nodes based on their spatial similarity. Then, these subgraphs are sequentially connected by their root nodes and merged into a complete graph. Based on this graph, we devise a graph convolutional network for textual reasoning (GTR) by supervising it with a cross-entropy loss. GTR can be easily plugged in representative STR models to improve their performance owing to better textual reasoning. Specifically, we construct our model, namely S-GTR, by paralleling GTR to the language model in a segmentation-based STR baseline, which can effectively exploit the visual-linguistic complementarity via mutual learning. S-GTR sets new state-of-the-art on six challenging STR benchmarks and generalizes well to multi-linguistic datasets. Code is available at https://github.com/adeline-cs/GTR.