Large-scale diffusion models have achieved state-of-the-art results on text-to-image synthesis (T2I) tasks. Despite their ability to generate high-quality yet creative images, we observe that attribution-binding and compositional capabilities are still considered major challenging issues, especially when involving multiple objects. In this work, we improve the compositional skills of T2I models, specifically more accurate attribute binding and better image compositions. To do this, we incorporate linguistic structures with the diffusion guidance process based on the controllable properties of manipulating cross-attention layers in diffusion-based T2I models. We observe that keys and values in cross-attention layers have strong semantic meanings associated with object layouts and content. Therefore, we can better preserve the compositional semantics in the generated image by manipulating the cross-attention representations based on linguistic insights. Built upon Stable Diffusion, a SOTA T2I model, our structured cross-attention design is efficient that requires no additional training samples. We achieve better compositional skills in qualitative and quantitative results, leading to a 5-8% advantage in head-to-head user comparison studies. Lastly, we conduct an in-depth analysis to reveal potential causes of incorrect image compositions and justify the properties of cross-attention layers in the generation process.

Prompt tuning is a new few-shot transfer learning technique that only tunes the learnable prompt for pre-trained vision and language models such as CLIP. However, existing prompt tuning methods tend to learn spurious or entangled representations, which leads to poor generalization to unseen concepts. Towards non-spurious and efficient prompt learning from limited examples, this paper presents a novel \underline{\textbf{C}}ounterfactual \underline{\textbf{P}}rompt \underline{\textbf{L}}earning (CPL) method for vision and language models, which simultaneously employs counterfactual generation and contrastive learning in a joint optimization framework. Particularly, CPL constructs counterfactual by identifying minimal non-spurious feature change between semantically-similar positive and negative samples that causes concept change, and learns more generalizable prompt representation from both factual and counterfactual examples via contrastive learning. Extensive experiments demonstrate that CPL can obtain superior few-shot performance on different vision and language tasks than previous prompt tuning methods on CLIP. On image classification, we achieve 3.55\% average relative improvement on unseen classes across seven datasets; on image-text retrieval and visual question answering, we gain up to 4.09\% and 25.08\% relative improvements across three few-shot scenarios on unseen test sets respectively.

建立一个对话体现的代理执行现实生活任务一直是一个长期而又具有挑战性的研究目标，因为它需要有效的人类代理沟通，多模式理解，远程顺序决策等。传统的符号方法具有扩展和概括问题，而端到端的深度学习模型则遭受数据稀缺和高任务复杂性的影响，并且通常很难解释。为了从两全其美的世界中受益，我们提出了一个神经符号常识性推理（JARVIS）框架，用于模块化，可推广和可解释的对话体现的药物。首先，它通过提示大型语言模型（LLM）来获得符号表示，以了解语言理解和次目标计划，并通过从视觉观察中构建语义图。然后，基于任务和动作级别的常识，次目标计划和行动生成的符号模块。在Teach数据集上进行的大量实验验证了我们的JARVIS框架的功效和效率，该框架在所有三个基于对话框的具体任务上实现了最新的（SOTA）结果，包括对话记录（EDH）的执行，对话框的轨迹， （TFD）和两个代理任务完成（TATC）（例如，我们的方法将EDH看不见的成功率从6.1 \％\％提高到15.8 \％）。此外，我们系统地分析了影响任务绩效的基本因素，并在几个射击设置中证明了我们方法的优越性。我们的Jarvis模型在Alexa奖Simbot公共基准挑战赛中排名第一。

In computer vision, it has achieved great transfer learning performance via adapting large-scale pretrained vision models (e.g., vision transformers) to downstream tasks. Common approaches for model adaptation either update all model parameters or leverage linear probes. In this paper, we aim to study parameter-efficient model adaptation strategies for vision transformers on the image classification task. We formulate efficient model adaptation as a subspace training problem and perform a comprehensive benchmarking over different efficient adaptation methods. We conduct an empirical study on each efficient model adaptation method focusing on its performance alongside parameter cost. Furthermore, we propose a parameter-efficient model adaptation framework, which first selects submodules by measuring local intrinsic dimensions and then projects them into subspace for further decomposition via a novel Kronecker Adaptation (KAdaptation) method. We analyze and compare our method with a diverse set of baseline model adaptation methods (including state-of-the-art methods for pretrained language models). Our method performs the best in terms of the tradeoff between accuracy and parameter efficiency across 20 image classification datasets under the few-shot setting and 7 image classification datasets under the full-shot setting.

Designing experiments often requires balancing between learning about the true treatment effects and earning from allocating more samples to the superior treatment. While optimal algorithms for the Multi-Armed Bandit Problem (MABP) provide allocation policies that optimally balance learning and earning, they tend to be computationally expensive. The Gittins Index (GI) is a solution to the MABP that can simultaneously attain optimality and computationally efficiency goals, and it has been recently used in experiments with Bernoulli and Gaussian rewards. For the first time, we present a modification of the GI rule that can be used in experiments with exponentially-distributed rewards. We report its performance in simulated 2- armed and 3-armed experiments. Compared to traditional non-adaptive designs, our novel GI modified design shows operating characteristics comparable in learning (e.g. statistical power) but substantially better in earning (e.g. direct benefits). This illustrates the potential that designs using a GI approach to allocate participants have to improve participant benefits, increase efficiencies, and reduce experimental costs in adaptive multi-armed experiments with exponential rewards.

Transformer has achieved impressive successes for various computer vision tasks. However, most of existing studies require to pretrain the Transformer backbone on a large-scale labeled dataset (e.g., ImageNet) for achieving satisfactory performance, which is usually unavailable for medical images. Additionally, due to the gap between medical and natural images, the improvement generated by the ImageNet pretrained weights significantly degrades while transferring the weights to medical image processing tasks. In this paper, we propose Bootstrap Own Latent of Transformer (BOLT), a self-supervised learning approach specifically for medical image classification with the Transformer backbone. Our BOLT consists of two networks, namely online and target branches, for self-supervised representation learning. Concretely, the online network is trained to predict the target network representation of the same patch embedding tokens with a different perturbation. To maximally excavate the impact of Transformer from limited medical data, we propose an auxiliary difficulty ranking task. The Transformer is enforced to identify which branch (i.e., online/target) is processing the more difficult perturbed tokens. Overall, the Transformer endeavours itself to distill the transformation-invariant features from the perturbed tokens to simultaneously achieve difficulty measurement and maintain the consistency of self-supervised representations. The proposed BOLT is evaluated on three medical image processing tasks, i.e., skin lesion classification, knee fatigue fracture grading and diabetic retinopathy grading. The experimental results validate the superiority of our BOLT for medical image classification, compared to ImageNet pretrained weights and state-of-the-art self-supervised learning approaches.

Text clustering and topic extraction are two important tasks in text mining. Usually, these two tasks are performed separately. For topic extraction to facilitate clustering, we can first project texts into a topic space and then perform a clustering algorithm to obtain clusters. To promote topic extraction by clustering, we can first obtain clusters with a clustering algorithm and then extract cluster-specific topics. However, this naive strategy ignores the fact that text clustering and topic extraction are strongly correlated and follow a chicken-and-egg relationship. Performing them separately fails to make them mutually benefit each other to achieve the best overall performance. In this paper, we propose an unsupervised text clustering and topic extraction framework (ClusTop) which integrates text clustering and topic extraction into a unified framework and can achieve high-quality clustering result and extract topics from each cluster simultaneously. Our framework includes four components: enhanced language model training, dimensionality reduction, clustering and topic extraction, where the enhanced language model can be viewed as a bridge between clustering and topic extraction. On one hand, it provides text embeddings with a strong cluster structure which facilitates effective text clustering; on the other hand, it pays high attention on the topic related words for topic extraction because of its self-attention architecture. Moreover, the training of enhanced language model is unsupervised. Experiments on two datasets demonstrate the effectiveness of our framework and provide benchmarks for different model combinations in this framework.

This paper illustrates the technologies of user next intent prediction with a concept knowledge graph. The system has been deployed on the Web at Alipay, serving more than 100 million daily active users. Specifically, we propose AlipayKG to explicitly characterize user intent, which is an offline concept knowledge graph in the Life-Service domain modeling the historical behaviors of users, the rich content interacted by users and the relations between them. We further introduce a Transformer-based model which integrates expert rules from the knowledge graph to infer the online user's next intent. Experimental results demonstrate that the proposed system can effectively enhance the performance of the downstream tasks while retaining explainability.

Capturing feature information effectively is of great importance in vision tasks. With the development of convolutional neural networks (CNNs), concepts like residual connection and multiple scales promote continual performance gains on diverse deep learning vision tasks. However, the existing methods do not organically combined advantages of these valid ideas. In this paper, we propose a novel CNN architecture called GoogLe2Net, it consists of residual feature-reutilization inceptions (ResFRI) or split residual feature-reutilization inceptions (Split-ResFRI) which create transverse passages between adjacent groups of convolutional layers to enable features flow to latter processing branches and possess residual connections to better process information. Our GoogLe2Net is able to reutilize information captured by foregoing groups of convolutional layers and express multi-scale features at a fine-grained level, which improves performances in image classification. And the inception we proposed could be embedded into inception-like networks directly without any migration costs. Moreover, in experiments based on popular vision datasets, such as CIFAR10 (97.94%), CIFAR100 (85.91%) and Tiny Imagenet (70.54%), we obtain better results on image classification task compared with other modern models.

Despite some successful applications of goal-driven navigation, existing deep reinforcement learning-based approaches notoriously suffers from poor data efficiency issue. One of the reasons is that the goal information is decoupled from the perception module and directly introduced as a condition of decision-making, resulting in the goal-irrelevant features of the scene representation playing an adversary role during the learning process. In light of this, we present a novel Goal-guided Transformer-enabled reinforcement learning (GTRL) approach by considering the physical goal states as an input of the scene encoder for guiding the scene representation to couple with the goal information and realizing efficient autonomous navigation. More specifically, we propose a novel variant of the Vision Transformer as the backbone of the perception system, namely Goal-guided Transformer (GoT), and pre-train it with expert priors to boost the data efficiency. Subsequently, a reinforcement learning algorithm is instantiated for the decision-making system, taking the goal-oriented scene representation from the GoT as the input and generating decision commands. As a result, our approach motivates the scene representation to concentrate mainly on goal-relevant features, which substantially enhances the data efficiency of the DRL learning process, leading to superior navigation performance. Both simulation and real-world experimental results manifest the superiority of our approach in terms of data efficiency, performance, robustness, and sim-to-real generalization, compared with other state-of-art baselines. Demonstration videos are available at \colorb{https://youtu.be/93LGlGvaN0c.