With the attention mechanism, transformers achieve significant empirical successes. Despite the intuitive understanding that transformers perform relational inference over long sequences to produce desirable representations, we lack a rigorous theory on how the attention mechanism achieves it. In particular, several intriguing questions remain open: (a) What makes a desirable representation? (b) How does the attention mechanism infer the desirable representation within the forward pass? (c) How does a pretraining procedure learn to infer the desirable representation through the backward pass? We observe that, as is the case in BERT and ViT, input tokens are often exchangeable since they already include positional encodings. The notion of exchangeability induces a latent variable model that is invariant to input sizes, which enables our theoretical analysis. - To answer (a) on representation, we establish the existence of a sufficient and minimal representation of input tokens. In particular, such a representation instantiates the posterior distribution of the latent variable given input tokens, which plays a central role in predicting output labels and solving downstream tasks. - To answer (b) on inference, we prove that attention with the desired parameter infers the latent posterior up to an approximation error, which is decreasing in input sizes. In detail, we quantify how attention approximates the conditional mean of the value given the key, which characterizes how it performs relational inference over long sequences. - To answer (c) on learning, we prove that both supervised and self-supervised objectives allow empirical risk minimization to learn the desired parameter up to a generalization error, which is independent of input sizes. Particularly, in the self-supervised setting, we identify a condition number that is pivotal to solving downstream tasks.

One of the key challenges in deploying RL to real-world applications is to adapt to variations of unknown environment contexts, such as changing terrains in robotic tasks and fluctuated bandwidth in congestion control. Existing works on adaptation to unknown environment contexts either assume the contexts are the same for the whole episode or assume the context variables are Markovian. However, in many real-world applications, the environment context usually stays stable for a stochastic period and then changes in an abrupt and unpredictable manner within an episode, resulting in a segment structure, which existing works fail to address. To leverage the segment structure of piecewise stable context in real-world applications, in this paper, we propose a \textit{\textbf{Se}gmented \textbf{C}ontext \textbf{B}elief \textbf{A}ugmented \textbf{D}eep~(SeCBAD)} RL method. Our method can jointly infer the belief distribution over latent context with the posterior over segment length and perform more accurate belief context inference with observed data within the current context segment. The inferred belief context can be leveraged to augment the state, leading to a policy that can adapt to abrupt variations in context. We demonstrate empirically that SeCBAD can infer context segment length accurately and outperform existing methods on a toy grid world environment and Mujuco tasks with piecewise-stable context.

Video super-resolution is one of the most popular tasks on mobile devices, being widely used for an automatic improvement of low-bitrate and low-resolution video streams. While numerous solutions have been proposed for this problem, they are usually quite computationally demanding, demonstrating low FPS rates and power efficiency on mobile devices. In this Mobile AI challenge, we address this problem and propose the participants to design an end-to-end real-time video super-resolution solution for mobile NPUs optimized for low energy consumption. The participants were provided with the REDS training dataset containing video sequences for a 4X video upscaling task. The runtime and power efficiency of all models was evaluated on the powerful MediaTek Dimensity 9000 platform with a dedicated AI processing unit capable of accelerating floating-point and quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 500 FPS rate and 0.2 [Watt / 30 FPS] power consumption. A detailed description of all models developed in the challenge is provided in this paper.

现有的二进制神经网络（BNN）主要在具有二进制功能的局部卷积上运作。但是，这种简单的位操作缺乏建模上下文依赖性的能力，这对于学习视觉模型中的歧视性深度表示至关重要。在这项工作中，我们通过介绍二进制神经模块的新设计来解决这个问题，这使BNN能够学习有效的上下文依赖性。首先，我们建议二进制多层感知器（MLP）块作为二进制卷积块的替代方案，以直接建模上下文依赖性。短距离和远程特征依赖性均由二进制MLP建模，其中前者提供局部电感偏置，后者在二元卷积中有限的接受场有限。其次，为了提高具有上下文依赖性的二进制模型的鲁棒性，我们计算上下文动态嵌入，以确定一般二进制卷积块中的二进化阈值。用我们的二进制MLP块和改进的二进制卷积，我们用明确的上下文依赖性建模构建了BNN，称为BCDNET。在标准Imagenet-1K分类基准上，BCDNET可实现72.3％的TOP-1准确性，并且优于领先的二进制方法的差距很大。尤其是，提出的BCDNET超过了最新的ReactNet-A，具有相似操作的2.9％TOP-1准确性。我们的代码可从https://github.com/sense-gvt/bcdn获得

在这项工作中，我们提出了一个单视手网格重建框架，可以同时实现高重建精度，快速推断速度和时间相干性。具体而言，对于2D编码，我们提出了轻量级但有效的堆叠结构。关于3D解码，我们提供有效的图形操作员，即深度可分离的螺旋卷积。此外，我们提出了一种用于桥接2D和3D表示之间的间隙的新颖特征提升模块。该模块以基于地图的位置回归（MapReg）块开头，以集成HeatMap编码和位置回归范例的优点，以提高2D精度和时间相干性。此外，MapReg之后是姿势池和姿势到顶点提升方法，它将2D姿势编码转换为3D顶点的语义特征。总体而言，我们的手部重建框架称为MobRecon，包括经济实惠的计算成本和微型模型大小，在Apple A14 CPU上达到83FP的高推理速度。广泛的对流行数据集如弗里安，RHD和HO3DV2的实验表明，我们的Mobrecon在重建准确性和时间一致性方面取得了卓越的性能。我们的代码在https://github.com/seanchenxy/handmesh公开提供。

Knowledge graph embedding (KGE), which maps entities and relations in a knowledge graph into continuous vector spaces, has achieved great success in predicting missing links in knowledge graphs. However, knowledge graphs often contain incomplete triples that are difficult to inductively infer by KGEs. To address this challenge, we resort to analogical inference and propose a novel and general self-supervised framework AnKGE to enhance KGE models with analogical inference capability. We propose an analogical object retriever that retrieves appropriate analogical objects from entity-level, relation-level, and triple-level. And in AnKGE, we train an analogy function for each level of analogical inference with the original element embedding from a well-trained KGE model as input, which outputs the analogical object embedding. In order to combine inductive inference capability from the original KGE model and analogical inference capability enhanced by AnKGE, we interpolate the analogy score with the base model score and introduce the adaptive weights in the score function for prediction. Through extensive experiments on FB15k-237 and WN18RR datasets, we show that AnKGE achieves competitive results on link prediction task and well performs analogical inference.

As an important variant of entity alignment (EA), multi-modal entity alignment (MMEA) aims to discover identical entities across different knowledge graphs (KGs) with multiple modalities like images. However, current MMEA algorithms all adopt KG-level modality fusion strategies but ignore modality differences among individual entities, hurting the robustness to potential noise involved in modalities (e.g., unidentifiable images and relations). In this paper we present MEAformer, a multi-modal entity alignment transformer approach for meta modality hybrid, to dynamically predict the mutual correlation coefficients among modalities for instance-level feature fusion. A modal-aware hard entity replay strategy is also proposed for addressing vague entity details. Extensive experimental results show that our model not only achieves SOTA performance on multiple training scenarios including supervised, unsupervised, iterative, and low resource, but also has limited parameters, optimistic speed, and good interpretability. Our code will be available soon.

The ability to create realistic, animatable and relightable head avatars from casual video sequences would open up wide ranging applications in communication and entertainment. Current methods either build on explicit 3D morphable meshes (3DMM) or exploit neural implicit representations. The former are limited by fixed topology, while the latter are non-trivial to deform and inefficient to render. Furthermore, existing approaches entangle lighting in the color estimation, thus they are limited in re-rendering the avatar in new environments. In contrast, we propose PointAvatar, a deformable point-based representation that disentangles the source color into intrinsic albedo and normal-dependent shading. We demonstrate that PointAvatar bridges the gap between existing mesh- and implicit representations, combining high-quality geometry and appearance with topological flexibility, ease of deformation and rendering efficiency. We show that our method is able to generate animatable 3D avatars using monocular videos from multiple sources including hand-held smartphones, laptop webcams and internet videos, achieving state-of-the-art quality in challenging cases where previous methods fail, e.g., thin hair strands, while being significantly more efficient in training than competing methods.

The goal of multimodal abstractive summarization (MAS) is to produce a concise summary given the multimodal data (text and vision). Existing studies on MAS mainly focus on how to effectively use the extracted visual features, having achieved impressive success on the high-resource English dataset. However, less attention has been paid to the quality of the visual features to the summary, which may limit the model performance especially in the low- and zero-resource scenarios. In this paper, we propose to improve the summary quality through summary-oriented visual features. To this end, we devise two auxiliary tasks including \emph{vision to summary task} and \emph{masked image modeling task}. Together with the main summarization task, we optimize the MAS model via the training objectives of all these tasks. By these means, the MAS model can be enhanced by capturing the summary-oriented visual features, thereby yielding more accurate summaries. Experiments on 44 languages, covering mid-high-, low-, and zero-resource scenarios, verify the effectiveness and superiority of the proposed approach, which achieves state-of-the-art performance under all scenarios.

Neural Radiance Fields (NeRF) methods have proved effective as compact, high-quality and versatile representations for 3D scenes, and enable downstream tasks such as editing, retrieval, navigation, etc. Various neural architectures are vying for the core structure of NeRF, including the plain Multi-Layer Perceptron (MLP), sparse tensors, low-rank tensors, hashtables and their compositions. Each of these representations has its particular set of trade-offs. For example, the hashtable-based representations admit faster training and rendering but their lack of clear geometric meaning hampers downstream tasks like spatial-relation-aware editing. In this paper, we propose Progressive Volume Distillation (PVD), a systematic distillation method that allows any-to-any conversions between different architectures, including MLP, sparse or low-rank tensors, hashtables and their compositions. PVD consequently empowers downstream applications to optimally adapt the neural representations for the task at hand in a post hoc fashion. The conversions are fast, as distillation is progressively performed on different levels of volume representations, from shallower to deeper. We also employ special treatment of density to deal with its specific numerical instability problem. Empirical evidence is presented to validate our method on the NeRF-Synthetic, LLFF and TanksAndTemples datasets. For example, with PVD, an MLP-based NeRF model can be distilled from a hashtable-based Instant-NGP model at a 10X~20X faster speed than being trained the original NeRF from scratch, while achieving a superior level of synthesis quality. Code is available at https://github.com/megvii-research/AAAI2023-PVD.