Prompt tuning is a parameter-efficient method, which freezes all PLM parameters and only prepends some additional tunable tokens called soft prompts to the input text. However, soft prompts heavily rely on a better initialization and may easily result in overfitting under few-shot settings, which causes prompt-tuning performing much worse than fine-tuning. To address the above issues, this paper proposes a novel Self-sUpervised Meta-prompt learning framework with MEtagradient Regularization for few shot generalization (SUMMER). We leverage self-supervised meta-learning to better initialize soft prompts and curriculum-based task augmentation is further proposed to enrich the meta-task distribution. Besides, a novel meta-gradient regularization method is integrated into the meta-prompt learning framework, which meta-learns to transform the raw gradient during few-shot learning into a domain-generalizable direction, thus alleviating the problem of overfitting. Extensive experiments show that SUMMER achieves better performance for different few-shot downstream tasks, and also exhibits a stronger domain generalization ability.
Stereo image super-resolution aims to boost the performance of image super-resolution by exploiting the supplementary information provided by binocular systems. Although previous methods have achieved promising results, they did not fully utilize the information of cross-view and intra-view. To further unleash the potential of binocular images, in this letter, we propose a novel Transformerbased parallax fusion module called Parallax Fusion Transformer (PFT). PFT employs a Cross-view Fusion Transformer (CVFT) to utilize cross-view information and an Intra-view Refinement Transformer (IVRT) for intra-view feature refinement. Meanwhile, we adopted the Swin Transformer as the backbone for feature extraction and SR reconstruction to form a pure Transformer architecture called PFT-SSR. Extensive experiments and ablation studies show that PFT-SSR achieves competitive results and outperforms most SOTA methods. Source code is available at https://github.com/MIVRC/PFT-PyTorch.
Scene Graph Generation (SGG) aims to extract <subject, predicate, object> relationships in images for vision understanding. Although recent works have made steady progress on SGG, they still suffer long-tail distribution issues that tail-predicates are more costly to train and hard to distinguish due to a small amount of annotated data compared to frequent predicates. Existing re-balancing strategies try to haddle it via prior rules but are still confined to pre-defined conditions, which are not scalable for various models and datasets. In this paper, we propose a Cross-modal prediCate boosting (CaCao) framework, where a visually-prompted language model is learned to generate diverse fine-grained predicates in a low-resource way. The proposed CaCao can be applied in a plug-and-play fashion and automatically strengthen existing SGG to tackle the long-tailed problem. Based on that, we further introduce a novel Entangled cross-modal prompt approach for open-world predicate scene graph generation (Epic), where models can generalize to unseen predicates in a zero-shot manner. Comprehensive experiments on three benchmark datasets show that CaCao consistently boosts the performance of multiple scene graph generation models in a model-agnostic way. Moreover, our Epic achieves competitive performance on open-world predicate prediction.
Prompt tuning is a parameter-efficient method, which learns soft prompts and conditions frozen language models to perform specific downstream tasks. Though effective, prompt tuning under few-shot settings on the one hand heavily relies on a good initialization of soft prompts. On the other hand, it can easily result in overfitting. Existing works leverage pre-training or supervised meta-learning to initialize soft prompts but they cannot data-efficiently generalize to unseen downstream tasks. To address the above problems, this paper proposes a novel Self-sUpervised meta-Prompt learning framework with meta-gradient Regularization for few-shot generalization (SUPMER). We first design a set of self-supervised anchor meta-training tasks with different task formats and further enrich the task distribution with curriculum-based task augmentation. Then a novel meta-gradient regularization method is integrated into meta-prompt learning. It meta-learns to transform the raw gradients during few-shot learning into a domain-generalizable direction, thus alleviating the problem of overfitting. Extensive experiments show that SUPMER achieves better performance for different few-shot downstream tasks, and also exhibits a stronger domain generalization ability.
Prompt tuning, a recently emerging paradigm, enables the powerful vision-language pre-training models to adapt to downstream tasks in a parameter -- and data -- efficient way, by learning the ``soft prompts'' to condition frozen pre-training models. Though effective, it is particularly problematic in the few-shot scenario, where prompt tuning performance is sensitive to the initialization and requires a time-consuming process to find a good initialization, thus restricting the fast adaptation ability of the pre-training models. In addition, prompt tuning could undermine the generalizability of the pre-training models, because the learnable prompt tokens are easy to overfit to the limited training samples. To address these issues, we introduce a novel Gradient-RegulAted Meta-prompt learning (GRAM) framework that jointly meta-learns an efficient soft prompt initialization for better adaptation and a lightweight gradient regulating function for strong cross-domain generalizability in a meta-learning paradigm using only the unlabeled image-text pre-training data. Rather than designing a specific prompt tuning method, our GRAM can be easily incorporated into various prompt tuning methods in a model-agnostic way, and comprehensive experiments show that GRAM brings about consistent improvement for them in several settings (i.e., few-shot learning, cross-domain generalization, cross-dataset generalization, etc.) over 11 datasets. Further, experiments show that GRAM enables the orthogonal methods of textual and visual prompt tuning to work in a mutually-enhanced way, offering better generalizability beyond the uni-modal prompt tuning methods.
In the past decade, convolutional neural networks (CNNs) have shown prominence for semantic segmentation. Although CNN models have very impressive performance, the ability to capture global representation is still insufficient, which results in suboptimal results. Recently, Transformer achieved huge success in NLP tasks, demonstrating its advantages in modeling long-range dependency. Recently, Transformer has also attracted tremendous attention from computer vision researchers who reformulate the image processing tasks as a sequence-to-sequence prediction but resulted in deteriorating local feature details. In this work, we propose a lightweight real-time semantic segmentation network called LETNet. LETNet combines a U-shaped CNN with Transformer effectively in a capsule embedding style to compensate for respective deficiencies. Meanwhile, the elaborately designed Lightweight Dilated Bottleneck (LDB) module and Feature Enhancement (FE) module cultivate a positive impact on training from scratch simultaneously. Extensive experiments performed on challenging datasets demonstrate that LETNet achieves superior performances in accuracy and efficiency balance. Specifically, It only contains 0.95M parameters and 13.6G FLOPs but yields 72.8\% mIoU at 120 FPS on the Cityscapes test set and 70.5\% mIoU at 250 FPS on the CamVid test dataset using a single RTX 3090 GPU. The source code will be available at https://github.com/IVIPLab/LETNet.
Temporal grounding is the task of locating a specific segment from an untrimmed video according to a query sentence. This task has achieved significant momentum in the computer vision community as it enables activity grounding beyond pre-defined activity classes by utilizing the semantic diversity of natural language descriptions. The semantic diversity is rooted in the principle of compositionality in linguistics, where novel semantics can be systematically described by combining known words in novel ways (compositional generalization). However, existing temporal grounding datasets are not carefully designed to evaluate the compositional generalizability. To systematically benchmark the compositional generalizability of temporal grounding models, we introduce a new Compositional Temporal Grounding task and construct two new dataset splits, i.e., Charades-CG and ActivityNet-CG. When evaluating the state-of-the-art methods on our new dataset splits, we empirically find that they fail to generalize to queries with novel combinations of seen words. We argue that the inherent structured semantics inside the videos and language is the crucial factor to achieve compositional generalization. Based on this insight, we propose a variational cross-graph reasoning framework that explicitly decomposes video and language into hierarchical semantic graphs, respectively, and learns fine-grained semantic correspondence between the two graphs. Furthermore, we introduce a novel adaptive structured semantics learning approach to derive the structure-informed and domain-generalizable graph representations, which facilitate the fine-grained semantic correspondence reasoning between the two graphs. Extensive experiments validate the superior compositional generalizability of our approach.
Recently, great progress has been made in single-image super-resolution (SISR) based on deep learning technology. However, the existing methods usually require a large computational cost. Meanwhile, the activation function will cause some features of the intermediate layer to be lost. Therefore, it is a challenge to make the model lightweight while reducing the impact of intermediate feature loss on the reconstruction quality. In this paper, we propose a Feature Interaction Weighted Hybrid Network (FIWHN) to alleviate the above problem. Specifically, FIWHN consists of a series of novel Wide-residual Distillation Interaction Blocks (WDIB) as the backbone, where every third WDIBs form a Feature shuffle Weighted Group (FSWG) by mutual information mixing and fusion. In addition, to mitigate the adverse effects of intermediate feature loss on the reconstruction results, we introduced a well-designed Wide Convolutional Residual Weighting (WCRW) and Wide Identical Residual Weighting (WIRW) units in WDIB, and effectively cross-fused features of different finenesses through a Wide-residual Distillation Connection (WRDC) framework and a Self-Calibrating Fusion (SCF) unit. Finally, to complement the global features lacking in the CNN model, we introduced the Transformer into our model and explored a new way of combining the CNN and Transformer. Extensive quantitative and qualitative experiments on low-level and high-level tasks show that our proposed FIWHN can achieve a good balance between performance and efficiency, and is more conducive to downstream tasks to solve problems in low-pixel scenarios.
Many studies have been conducted to improve the efficiency of Transformer from quadric to linear. Among them, the low-rank-based methods aim to learn the projection matrices to compress the sequence length. However, the projection matrices are fixed once they have been learned, which compress sequence length with dedicated coefficients for tokens in the same position. Adopting such input-invariant projections ignores the fact that the most informative part of a sequence varies from sequence to sequence, thus failing to preserve the most useful information that lies in varied positions. In addition, previous efficient Transformers only focus on the influence of sequence length while neglecting the effect of hidden state dimension. To address the aforementioned problems, we present an efficient yet effective attention mechanism, namely the Dynamic Bilinear Low-Rank Attention (DBA), which compresses the sequence length by input-sensitive dynamic projection matrices and achieves linear time and space complexity by jointly optimizing the sequence length and hidden state dimension while maintaining state-of-the-art performance. Specifically, we first theoretically demonstrate that the sequence length can be compressed non-destructively from a novel perspective of information theory, with compression matrices dynamically determined by the input sequence. Furthermore, we show that the hidden state dimension can be approximated by extending the Johnson-Lindenstrauss lemma, optimizing the attention in bilinear form. Theoretical analysis shows that DBA is proficient in capturing high-order relations in cross-attention problems. Experiments over tasks with diverse sequence length conditions show that DBA achieves state-of-the-art performance compared with various strong baselines while maintaining less memory consumption with higher speed.
Large-scale vision-language pre-training has shown impressive advances in a wide range of downstream tasks. Existing methods mainly model the cross-modal alignment by the similarity of the global representations of images and texts, or advanced cross-modal attention upon image and text features. However, they fail to explicitly learn the fine-grained semantic alignment between visual regions and textual phrases, as only global image-text alignment information is available. In this paper, we introduce LOUPE, a fine-grained semantically aLigned visiOn-langUage PrE-training framework, which learns fine-grained semantic alignment from the novel perspective of game-theoretic interactions. To efficiently compute the game-theoretic interactions, we further propose an uncertainty-aware neural Shapley interaction learning module. Experiments show that LOUPE achieves state-of-the-art on image-text retrieval benchmarks. Without any object-level human annotations and fine-tuning, LOUPE achieves competitive performance on object detection and visual grounding. More importantly, LOUPE opens a new promising direction of learning fine-grained semantics from large-scale raw image-text pairs.