Question Generation (QG) is a fundamental NLP task for many downstream applications. Recent studies on open-book QG, where supportive answer-context pairs are provided to models, have achieved promising progress. However, generating natural questions under a more practical closed-book setting that lacks these supporting documents still remains a challenge. In this work, we propose a new QG model for this closed-book setting that is designed to better understand the semantics of long-form abstractive answers and store more information in its parameters through contrastive learning and an answer reconstruction module. Through experiments, we validate the proposed QG model on both public datasets and a new WikiCQA dataset. Empirical results show that the proposed QG model outperforms baselines in both automatic evaluation and human evaluation. In addition, we show how to leverage the proposed model to improve existing question-answering systems. These results further indicate the effectiveness of our QG model for enhancing closed-book question-answering tasks.
How to achieve neural machine translation with limited parallel data? Existing techniques often rely on large-scale monolingual corpora, which is impractical for some low-resource languages. In this paper, we turn to connect several low-resource languages to a particular high-resource one by additional visual modality. Specifically, we propose a cross-modal contrastive learning method to learn a shared space for all languages, where both a coarse-grained sentence-level objective and a fine-grained token-level one are introduced. Experimental results and further analysis show that our method can effectively learn the cross-modal and cross-lingual alignment with a small amount of image-text pairs and achieves significant improvements over the text-only baseline under both zero-shot and few-shot scenarios.
Heterogeneous Face Recognition (HFR) refers to matching face images captured in different domains, such as thermal to visible images (VIS), sketches to visible images, near-infrared to visible, and so on. This is particularly useful in matching visible spectrum images to images captured from other modalities. Though highly useful, HFR is challenging because of the domain gap between the source and target domain. Often, large-scale paired heterogeneous face image datasets are absent, preventing training models specifically for the heterogeneous task. In this work, we propose a surprisingly simple, yet, very effective method for matching face images across different sensing modalities. The core idea of the proposed approach is to add a novel neural network block called Prepended Domain Transformer (PDT) in front of a pre-trained face recognition (FR) model to address the domain gap. Retraining this new block with few paired samples in a contrastive learning setup was enough to achieve state-of-the-art performance in many HFR benchmarks. The PDT blocks can be retrained for several source-target combinations using the proposed general framework. The proposed approach is architecture agnostic, meaning they can be added to any pre-trained FR models. Further, the approach is modular and the new block can be trained with a minimal set of paired samples, making it much easier for practical deployment. The source code and protocols will be made available publicly.
Video language pre-training methods have mainly adopted sparse sampling techniques to alleviate the temporal redundancy of videos. Though effective, sparse sampling still suffers inter-modal redundancy: visual redundancy and textual redundancy. Compared with highly generalized text, sparsely sampled frames usually contain text-independent portions, called visual redundancy. Sparse sampling is also likely to miss important frames corresponding to some text portions, resulting in textual redundancy. Inter-modal redundancy leads to a mismatch of video and text information, hindering the model from better learning the shared semantics across modalities. To alleviate it, we propose Redundancy-aware Video-language Pre-training. We design a redundancy measurement of video patches and text tokens by calculating the cross-modal minimum dis-similarity. Then, we penalize the highredundant video patches and text tokens through a proposed redundancy-aware contrastive learning. We evaluate our method on four benchmark datasets, MSRVTT, MSVD, DiDeMo, and LSMDC, achieving a significant improvement over the previous stateof-the-art results. Our code are available at https://github.com/caskcsg/VLP/tree/main/RaP.
Contrastive learning has been extensively studied in sentence embedding learning, which assumes that the embeddings of different views of the same sentence are closer. The constraint brought by this assumption is weak, and a good sentence representation should also be able to reconstruct the original sentence fragments. Therefore, this paper proposes an information-aggregated contrastive learning framework for learning unsupervised sentence embeddings, termed InfoCSE. InfoCSE forces the representation of [CLS] positions to aggregate denser sentence information by introducing an additional Masked language model task and a well-designed network. We evaluate the proposed InfoCSE on several benchmark datasets w.r.t the semantic text similarity (STS) task. Experimental results show that InfoCSE outperforms SimCSE by an average Spearman correlation of 2.60% on BERT-base, and 1.77% on BERT-large, achieving state-of-the-art results among unsupervised sentence representation learning methods. Our code are available at https://github.com/caskcsg/sentemb/tree/main/InfoCSE.
Humans have a unique ability to learn new representations from just a handful of examples with little to no supervision. Deep learning models, however, require an abundance of data and supervision to perform at a satisfactory level. Unsupervised few-shot learning (U-FSL) is the pursuit of bridging this gap between machines and humans. Inspired by the capacity of graph neural networks (GNNs) in discovering complex inter-sample relationships, we propose a novel self-attention based message passing contrastive learning approach (coined as SAMP-CLR) for U-FSL pre-training. We also propose an optimal transport (OT) based fine-tuning strategy (we call OpT-Tune) to efficiently induce task awareness into our novel end-to-end unsupervised few-shot classification framework (SAMPTransfer). Our extensive experimental results corroborate the efficacy of SAMPTransfer in a variety of downstream few-shot classification scenarios, setting a new state-of-the-art for U-FSL on both miniImagenet and tieredImagenet benchmarks, offering up to 7%+ and 5%+ improvements, respectively. Our further investigations also confirm that SAMPTransfer remains on-par with some supervised baselines on miniImagenet and outperforms all existing U-FSL baselines in a challenging cross-domain scenario. Our code can be found in our GitHub repository at https://github.com/ojss/SAMPTransfer/.
Learning medical visual representations directly from paired radiology reports has become an emerging topic in representation learning. However, existing medical image-text joint learning methods are limited by instance or local supervision analysis, ignoring disease-level semantic correspondences. In this paper, we present a novel Multi-Granularity Cross-modal Alignment (MGCA) framework for generalized medical visual representation learning by harnessing the naturally exhibited semantic correspondences between medical image and radiology reports at three different levels, i.e., pathological region-level, instance-level, and disease-level. Specifically, we first incorporate the instance-wise alignment module by maximizing the agreement between image-report pairs. Further, for token-wise alignment, we introduce a bidirectional cross-attention strategy to explicitly learn the matching between fine-grained visual tokens and text tokens, followed by contrastive learning to align them. More important, to leverage the high-level inter-subject relationship semantic (e.g., disease) correspondences, we design a novel cross-modal disease-level alignment paradigm to enforce the cross-modal cluster assignment consistency. Extensive experimental results on seven downstream medical image datasets covering image classification, object detection, and semantic segmentation tasks demonstrate the stable and superior performance of our framework.
Large-scale video-language pre-training has shown significant improvement in video-language understanding tasks. Previous studies of video-language pretraining mainly focus on short-form videos (i.e., within 30 seconds) and sentences, leaving long-form video-language pre-training rarely explored. Directly learning representation from long-form videos and language may benefit many long-form video-language understanding tasks. However, it is challenging due to the difficulty of modeling long-range relationships and the heavy computational burden caused by more frames. In this paper, we introduce a Long-Form VIdeo-LAnguage pre-training model (LF-VILA) and train it on a large-scale long-form video and paragraph dataset constructed from an existing public dataset. To effectively capture the rich temporal dynamics and to better align video and language in an efficient end-to-end manner, we introduce two novel designs in our LF-VILA model. We first propose a Multimodal Temporal Contrastive (MTC) loss to learn the temporal relation across different modalities by encouraging fine-grained alignment between long-form videos and paragraphs. Second, we propose a Hierarchical Temporal Window Attention (HTWA) mechanism to effectively capture long-range dependency while reducing computational cost in Transformer. We fine-tune the pre-trained LF-VILA model on seven downstream long-form video-language understanding tasks of paragraph-to-video retrieval and long-form video question-answering, and achieve new state-of-the-art performances. Specifically, our model achieves 16.1% relative improvement on ActivityNet paragraph-to-video retrieval task and 2.4% on How2QA task, respectively. We release our code, dataset, and pre-trained models at https://github.com/microsoft/XPretrain.
Trajectory similarity measures act as query predicates in trajectory databases, making them the key player in determining the query results. They also have a heavy impact on the query efficiency. An ideal measure should have the capability to accurately evaluate the similarity between any two trajectories in a very short amount of time. Towards this aim, we propose a contrastive learning-based trajectory modeling method named TrajCL. We present four trajectory augmentation methods and a novel dual-feature self-attention-based trajectory backbone encoder. The resultant model can jointly learn both the spatial and the structural patterns of trajectories. Our model does not involve any recurrent structures and thus has a high efficiency. Besides, our pre-trained backbone encoder can be fine-tuned towards other computationally expensive measures with minimal supervision data. Experimental results show that TrajCL is consistently and significantly more accurate than the state-of-the-art trajectory similarity measures. After fine-tuning, i.e., to serve as an estimator for heuristic measures, TrajCL can even outperform the state-of-the-art supervised method by up to 56% in the accuracy for processing trajectory similarity queries.
Data in real-world object detection often exhibits the long-tailed distribution. Existing solutions tackle this problem by mitigating the competition between the head and tail categories. However, due to the scarcity of training samples, tail categories are still unable to learn discriminative representations. Bringing more data into the training may alleviate the problem, but collecting instance-level annotations is an excruciating task. In contrast, image-level annotations are easily accessible but not fully exploited. In this paper, we propose a novel framework CLIS (multi-task Collaborative Learning with Image-level Supervision), which leverage image-level supervision to enhance the detection ability in a multi-task collaborative way. Specifically, there are an object detection task (consisting of an instance-classification task and a localization task) and an image-classification task in our framework, responsible for utilizing the two types of supervision. Different tasks are trained collaboratively by three key designs: (1) task-specialized sub-networks that learn specific representations of different tasks without feature entanglement. (2) a siamese sub-network for the image-classification task that shares its knowledge with the instance-classification task, resulting in feature enrichment of detectors. (3) a contrastive learning regularization that maintains representation consistency, bridging feature gaps of different supervision. Extensive experiments are conducted on the challenging LVIS dataset. Without sophisticated loss engineering, CLIS achieves an overall AP of 31.1 with 10.1 point improvement on tail categories, establishing a new state-of-the-art. Code will be at https://github.com/waveboo/CLIS.
We introduce a new task, named video corpus visual answer localization (VCVAL), which aims to locate the visual answer in a large collection of untrimmed instructional videos using a natural language question. This task requires a range of skills - the interaction between vision and language, video retrieval, passage comprehension, and visual answer localization. In this paper, we propose a cross-modal contrastive global-span (CCGS) method for the VCVAL, jointly training the video corpus retrieval and visual answer localization subtasks with the global-span matrix. We have reconstructed a dataset named MedVidCQA, on which the VCVAL task is benchmarked. Experimental results show that the proposed method outperforms other competitive methods both in the video corpus retrieval and visual answer localization subtasks. Most importantly, we perform detailed analyses on extensive experiments, paving a new path for understanding the instructional videos, which ushers in further research.
Multimodal semantic understanding often has to deal with uncertainty, which means the obtained messages tend to refer to multiple targets. Such uncertainty is problematic for our interpretation, including inter- and intra-modal uncertainty. Little effort has studied the modeling of this uncertainty, particularly in pre-training on unlabeled datasets and fine-tuning in task-specific downstream datasets. In this paper, we project the representations of all modalities as probabilistic distributions via a Probability Distribution Encoder (PDE) by utilizing sequence-level interactions. Compared to the existing deterministic methods, such uncertainty modeling can convey richer multimodal semantic information and more complex relationships. Furthermore, we integrate uncertainty modeling with popular pre-training frameworks and propose suitable pre-training tasks: Distribution-based Vision-Language Contrastive learning (D-VLC), Distribution-based Masked Language Modeling (D-MLM), and Distribution-based Image-Text Matching (D-ITM). The fine-tuned models are applied to challenging downstream tasks, including image-text retrieval, visual question answering, visual reasoning, and visual entailment, and achieve state-of-the-art results.
The prevalence and perniciousness of fake news have been a critical issue on the Internet, which stimulates the development of automatic fake news detection in turn. In this paper, we focus on evidence-based fake news detection, where several evidences are utilized to probe the veracity of news (i.e., a claim). Most previous methods first employ sequential models to embed the semantic information and then capture the claim-evidence interaction based on attention mechanisms. Despite their effectiveness, they still suffer from three weaknesses. Firstly, sequential models fail to integrate the relevant information that is scattered far apart in evidences. Secondly, they underestimate much redundant information in evidences may be useless or harmful. Thirdly, insufficient data utilization limits the separability and reliability of representations captured by the model. To solve these problems, we propose a unified Graph-based sEmantic structure mining framework with ConTRAstive Learning, namely GETRAL in short. Specifically, we first model claims and evidences as graph-structured data to capture the long-distance semantic dependency. Consequently, we reduce information redundancy by performing graph structure learning. Then the fine-grained semantic representations are fed into the claim-evidence interaction module for predictions. Finally, an adversarial contrastive learning module is applied to make full use of data and strengthen representation learning. Comprehensive experiments have demonstrated the superiority of GETRAL over the state-of-the-arts and validated the efficacy of semantic mining with graph structure and contrastive learning.
Contrastive learning has recently achieved compelling performance in unsupervised sentence representation. As an essential element, data augmentation protocols, however, have not been well explored. The pioneering work SimCSE resorting to a simple dropout mechanism (viewed as continuous augmentation) surprisingly dominates discrete augmentations such as cropping, word deletion, and synonym replacement as reported. To understand the underlying rationales, we revisit existing approaches and attempt to hypothesize the desiderata of reasonable data augmentation methods: balance of semantic consistency and expression diversity. We then develop three simple yet effective discrete sentence augmentation schemes: punctuation insertion, modal verbs, and double negation. They act as minimal noises at lexical level to produce diverse forms of sentences. Furthermore, standard negation is capitalized on to generate negative samples for alleviating feature suppression involved in contrastive learning. We experimented extensively with semantic textual similarity on diverse datasets. The results support the superiority of the proposed methods consistently. Our key code is available at https://github.com/Zhudongsheng75/SDA
Few-shot object detection (FSOD) is to detect objects with a few examples. However, existing FSOD methods do not consider hierarchical fine-grained category structures of objects that exist widely in real life. For example, animals are taxonomically classified into orders, families, genera and species etc. In this paper, we propose and solve a new problem called hierarchical few-shot object detection (Hi-FSOD), which aims to detect objects with hierarchical categories in the FSOD paradigm. To this end, on the one hand, we build the first large-scale and high-quality Hi-FSOD benchmark dataset HiFSOD-Bird, which contains 176,350 wild-bird images falling to 1,432 categories. All the categories are organized into a 4-level taxonomy, consisting of 32 orders, 132 families, 572 genera and 1,432 species. On the other hand, we propose the first Hi-FSOD method HiCLPL, where a hierarchical contrastive learning approach is developed to constrain the feature space so that the feature distribution of objects is consistent with the hierarchical taxonomy and the model's generalization power is strengthened. Meanwhile, a probabilistic loss is designed to enable the child nodes to correct the classification errors of their parent nodes in the taxonomy. Extensive experiments on the benchmark dataset HiFSOD-Bird show that our method HiCLPL outperforms the existing FSOD methods.
The success of most advanced facial expression recognition works relies heavily on large-scale annotated datasets. However, it poses great challenges in acquiring clean and consistent annotations for facial expression datasets. On the other hand, self-supervised contrastive learning has gained great popularity due to its simple yet effective instance discrimination training strategy, which can potentially circumvent the annotation issue. Nevertheless, there remain inherent disadvantages of instance-level discrimination, which are even more challenging when faced with complicated facial representations. In this paper, we revisit the use of self-supervised contrastive learning and explore three core strategies to enforce expression-specific representations and to minimize the interference from other facial attributes, such as identity and face styling. Experimental results show that our proposed method outperforms the current state-of-the-art self-supervised learning methods, in terms of both categorical and dimensional facial expression recognition tasks.
Domain adaptation has been widely adopted to transfer styles across multi-vendors and multi-centers, as well as to complement the missing modalities. In this challenge, we proposed an unsupervised domain adaptation framework for cross-modality vestibular schwannoma (VS) and cochlea segmentation and Koos grade prediction. We learn the shared representation from both ceT1 and hrT2 images and recover another modality from the latent representation, and we also utilize proxy tasks of VS segmentation and brain parcellation to restrict the consistency of image structures in domain adaptation. After generating missing modalities, the nnU-Net model is utilized for VS and cochlea segmentation, while a semi-supervised contrastive learning pre-train approach is employed to improve the model performance for Koos grade prediction. On CrossMoDA validation phase Leaderboard, our method received rank 4 in task1 with a mean Dice score of 0.8394 and rank 2 in task2 with Macro-Average Mean Square Error of 0.3941. Our code is available at https://github.com/fiy2W/cmda2022.superpolymerization.
The self-supervised ultrasound (US) video model pretraining can use a small amount of labeled data to achieve one of the most promising results on US diagnosis. However, it does not take full advantage of multi-level knowledge for learning deep neural networks (DNNs), and thus is difficult to learn transferable feature representations. This work proposes a hierarchical contrastive learning (HiCo) method to improve the transferability for the US video model pretraining. HiCo introduces both peer-level semantic alignment and cross-level semantic alignment to facilitate the interaction between different semantic levels, which can effectively accelerate the convergence speed, leading to better generalization and adaptation of the learned model. Additionally, a softened objective function is implemented by smoothing the hard labels, which can alleviate the negative effect caused by local similarities of images between different classes. Experiments with HiCo on five datasets demonstrate its favorable results over state-of-the-art approaches. The source code of this work is publicly available at \url{https://github.com/983632847/HiCo}.
Recent methods for deep metric learning have been focusing on designing different contrastive loss functions between positive and negative pairs of samples so that the learned feature embedding is able to pull positive samples of the same class closer and push negative samples from different classes away from each other. In this work, we recognize that there is a significant semantic gap between features at the intermediate feature layer and class labels at the final output layer. To bridge this gap, we develop a contrastive Bayesian analysis to characterize and model the posterior probabilities of image labels conditioned by their features similarity in a contrastive learning setting. This contrastive Bayesian analysis leads to a new loss function for deep metric learning. To improve the generalization capability of the proposed method onto new classes, we further extend the contrastive Bayesian loss with a metric variance constraint. Our experimental results and ablation studies demonstrate that the proposed contrastive Bayesian metric learning method significantly improves the performance of deep metric learning in both supervised and pseudo-supervised scenarios, outperforming existing methods by a large margin.
Link prediction is the task of inferring missing links between entities in knowledge graphs. Embedding-based methods have shown effectiveness in addressing this problem by modeling relational patterns in triples. However, the link prediction task often requires contextual information in entity neighborhoods, while most existing embedding-based methods fail to capture it. Additionally, little attention is paid to the diversity of entity representations in different contexts, which often leads to false prediction results. In this situation, we consider that the schema of knowledge graph contains the specific contextual information, and it is beneficial for preserving the consistency of entities across contexts. In this paper, we propose a novel Schema-augmented Multi-level contrastive LEarning framework (SMiLE) to conduct knowledge graph link prediction. Specifically, we first exploit network schema as the prior constraint to sample negatives and pre-train our model by employing a multi-level contrastive learning method to yield both prior schema and contextual information. Then we fine-tune our model under the supervision of individual triples to learn subtler representations for link prediction. Extensive experimental results on four knowledge graph datasets with thorough analysis of each component demonstrate the effectiveness of our proposed framework against state-of-the-art baselines. The implementation of SMiLE is available at https://github.com/GKNL/SMiLE.