Narrative Incoherence Detection

Motivated by the increasing popularity of intelligent editing assistant, we introduce and investigate the task of narrative incoherence detection: Given a (corrupted) long-form narrative, decide whether there exists some semantic discrepancy in the narrative flow. Specifically, we focus on the missing sentence and incoherent sentence detection. Despite its simple setup, this task is challenging as the model needs to understand and analyze a multi-sentence narrative text, and make decisions at the sentence level. As an initial step towards this task, we implement several baselines either directly analyzing the raw text (\textit{token-level}) or analyzing learned sentence representations (\textit{sentence-level}). We observe that while token-level modeling enjoys greater expressive power and hence better performance, sentence-level modeling possesses an advantage in efficiency and flexibility. With pre-training on large-scale data and cycle-consistent sentence embedding, our extended sentence-level model can achieve comparable detection accuracy to the token-level model. As a by-product, such a strategy enables simultaneous incoherence detection and infilling/modification suggestions.

EasyTransfer -- A Simple and Scalable Deep Transfer Learning Platform for NLP Applications

The literature has witnessed the success of applying deep Transfer Learning (TL) algorithms to many NLP applications, yet it is not easy to build a simple and scalable TL toolkit for this purpose. To bridge this gap, the EasyTransfer platform is designed to make it easy to develop deep TL algorithms for NLP applications. It is built with rich API abstractions, a scalable architecture and comprehensive deep TL algorithms, to make the development of NLP applications easier. To be specific, the build-in data and model parallelism strategy shows to be 4x faster than the default distribution strategy of Tensorflow. EasyTransfer supports the mainstream pre-trained ModelZoo, including Pre-trained Language Models (PLMs) and multi-modality models. It also integrates various SOTA models for mainstream NLP applications in AppZoo, and supports mainstream TL algorithms as well. The toolkit is convenient for users to quickly start model training, evaluation, offline prediction, and online deployment. This system is currently deployed at Alibaba to support a variety of business scenarios, including item recommendation, personalized search, and conversational question answering. Extensive experiments on real-world datasets show that EasyTransfer is suitable for online production with cutting-edge performance. The source code of EasyTransfer is released at Github (https://github.com/alibaba/EasyTransfer).

Spherical Knowledge Distillation

Knowledge distillation aims at obtaining a small but effective deep model by transferring knowledge from a much larger one. The previous approaches try to reach this goal by simply "logit-supervised" information transferring between the teacher and student, which somehow can be subsequently decomposed as the transfer of normalized logits and $l^2$ norm. We argue that the norm of logits is actually interference, which damages the efficiency in the transfer process. To address this problem, we propose Spherical Knowledge Distillation (SKD). Specifically, we project the teacher and the student's logits into a unit sphere, and then we can efficiently perform knowledge distillation on the sphere. We verify our argument via theoretical analysis and ablation study. Extensive experiments have demonstrated the superiority and scalability of our method over the SOTAs.

Geometry-based Occlusion-Aware Unsupervised Stereo Matching for Autonomous Driving

Recently, there are emerging many stereo matching methods for autonomous driving based on unsupervised learning. Most of them take advantage of reconstruction losses to remove dependency on disparity groundtruth. Occlusion handling is a challenging problem in stereo matching, especially for unsupervised methods. Previous unsupervised methods failed to take full advantage of geometry properties in occlusion handling. In this paper, we introduce an effective way to detect occlusion regions and propose a novel unsupervised training strategy to deal with occlusion that only uses the predicted left disparity map, by making use of its geometry features in an iterative way. In the training process, we regard the predicted left disparity map as pseudo groundtruth and infer occluded regions using geometry features. The resulting occlusion mask is then used in either training, post-processing, or both of them as guidance. Experiments show that our method could deal with the occlusion problem effectively and significantly outperforms the other unsupervised methods for stereo matching. Moreover, our occlusion-aware strategies can be extended to the other stereo methods conveniently and improve their performances.

Accelerate Your CNN from Three Dimensions: A Comprehensive Pruning Framework

To deploy a pre-trained deep CNN on resource-constrained mobile devices, neural network pruning is often used to cut down the model's computational cost. For example, filter-level pruning (reducing the model's width) or layer-level pruning (reducing the model's depth) can both save computations with some sacrifice of accuracy. Besides, reducing the resolution of input images can also reach the same goal. Most previous methods focus on reducing one or two of these dimensions (i.e., depth, width, and image resolution) for acceleration. However, excessive reduction of any single dimension will lead to unacceptable accuracy loss, and we have to prune these three dimensions comprehensively to yield the best result. In this paper, a simple yet effective pruning framework is proposed to comprehensively consider these three dimensions. Our framework falls into two steps: 1) Determining the optimal depth (d*), width (w*), and image resolution (r) for the model. 2) Pruning the model in terms of (d*, w*, r*). Specifically, at the first step, we formulate model acceleration as an optimization problem. It takes depth (d), width (w) and image resolution (r) as variables and the model's accuracy as the optimization objective. Although it is hard to determine the expression of the objective function, approximating it with polynomials is still feasible, during which several properties of the objective function are utilized to ease and speedup the fitting process. Then the optimal d*, w* and r* are attained by maximizing the objective function with Lagrange multiplier theorem and KKT conditions. Extensive experiments are done on several popular architectures and datasets. The results show that we have outperformd the state-of-the-art pruning methods. The code will be published soon.

Does Network Width Really Help Adversarial Robustness?

Adversarial training is currently the most powerful defense against adversarial examples. Previous empirical results suggest that adversarial training requires wider networks for better performances. Yet, it remains elusive how does neural network width affects model robustness. In this paper, we carefully examine the relation between network width and model robustness. We present an intriguing phenomenon that the increased network width may not help robustness. Specifically, we show that the model robustness is closely related to both natural accuracy and perturbation stability, a new metric proposed in our paper to characterize the model's stability under adversarial perturbations. While better natural accuracy can be achieved on wider neural networks, the perturbation stability actually becomes worse, leading to a potentially worse overall model robustness. To understand the origin of this phenomenon, we further relate the perturbation stability with the network's local Lipschitznesss. By leveraging recent results on neural tangent kernels, we show that larger network width naturally leads to worse perturbation stability. This suggests that to fully unleash the power of wide model architecture, practitioners should adopt a larger regularization parameter for training wider networks. Experiments on benchmark datasets confirm that this strategy could indeed alleviate the perturbation stability issue and improve the state-of-the-art robust models.

Interpretable Real-Time Win Prediction for Honor of Kings, a Popular Mobile MOBA Esport

With the rapid prevalence and explosive development of MOBA esports (Multiplayer Online Battle Arena electronic sports), many research efforts have been devoted to automatically predicting the game results (win predictions). While this task has great potential in various applications such as esports live streaming and game commentator AI systems, previous studies suffer from two major limitations: 1) insufficient real-time input features and high-quality training data; 2) non-interpretable inference processes of the black-box prediction models. To mitigate these issues, we collect and release a large-scale dataset that contains real-time game records with rich input features of the popular MOBA game Honor of Kings. For interpretable predictions, we propose a Two-Stage Spatial-Temporal Network (TSSTN) that can not only provide accurate real-time win predictions but also attribute the ultimate prediction results to the contributions of different features for interpretability. Experiment results and applications in real-world live streaming scenarios show that the proposed TSSTN model is effective both in prediction accuracy and interpretability.

RESA: Recurrent Feature-Shift Aggregator for Lane Detection

Lane detection is one of the most important tasks in self-driving. Due to various complex scenarios (e.g., severe occlusion, ambiguous lanes, and etc.) and the sparse supervisory signals inherent in lane annotations, lane detection task is still challenging. Thus, it is difficult for ordinary convolutional neural network (CNN) trained in general scenes to catch subtle lane feature from raw image. In this paper, we present a novel module named REcurrent Feature-Shift Aggregator (RESA) to enrich lane feature after preliminary feature extraction with an ordinary CNN. RESA takes advantage of strong shape priors of lanes and captures spatial relationships of pixels across rows and columns. It shifts sliced feature map recurrently in vertical and horizontal directions and enables each pixel to gather global information. With the help of slice-by-slice information propagation, RESA can conjecture lanes accurately in challenging scenarios with weak appearance clues. Moreover, we also propose a Bilateral Up-Sampling Decoder which combines coarse grained feature and fine detailed feature in up-sampling stage, and it can recover low-resolution feature map into pixel-wise prediction meticulously. Our method achieves state-of-the-art results on two popular lane detection benchmarks (CULane and Tusimple). The code will be released publicly available.

Apparel-invariant Feature Learning for Apparel-changed Person Re-identification

With the rise of deep learning methods, person Re-Identification (ReID) performance has been improved tremendously in many public datasets. However, most public ReID datasets are collected in a short time window in which persons' appearance rarely changes. In real-world applications such as in a shopping mall, the same person's clothing may change, and different persons may wearing similar clothes. All these cases can result in an inconsistent ReID performance, revealing a critical problem that current ReID models heavily rely on person's apparels. Therefore, it is critical to learn an apparel-invariant person representation under cases like cloth changing or several persons wearing similar clothes. In this work, we tackle this problem from the viewpoint of invariant feature representation learning. The main contributions of this work are as follows. (1) We propose the semi-supervised Apparel-invariant Feature Learning (AIFL) framework to learn an apparel-invariant pedestrian representation using images of the same person wearing different clothes. (2) To obtain images of the same person wearing different clothes, we propose an unsupervised apparel-simulation GAN (AS-GAN) to synthesize cloth changing images according to the target cloth embedding. It's worth noting that the images used in ReID tasks were cropped from real-world low-quality CCTV videos, making it more challenging to synthesize cloth changing images. We conduct extensive experiments on several datasets comparing with several baselines. Experimental results demonstrate that our proposal can improve the ReID performance of the baseline models.

Learning to Caricature via Semantic Shape Transform

Caricature is an artistic drawing created to abstract or exaggerate facial features of a person. Rendering visually pleasing caricatures is a difficult task that requires professional skills, and thus it is of great interest to design a method to automatically generate such drawings. To deal with large shape changes, we propose an algorithm based on a semantic shape transform to produce diverse and plausible shape exaggerations. Specifically, we predict pixel-wise semantic correspondences and perform image warping on the input photo to achieve dense shape transformation. We show that the proposed framework is able to render visually pleasing shape exaggerations while maintaining their facial structures. In addition, our model allows users to manipulate the shape via the semantic map. We demonstrate the effectiveness of our approach on a large photograph-caricature benchmark dataset with comparisons to the state-of-the-art methods.