With the success of Vision Transformers (ViTs) in computer vision tasks, recent arts try to optimize the performance and complexity of ViTs to enable efficient deployment on mobile devices. Multiple approaches are proposed to accelerate attention mechanism, improve inefficient designs, or incorporate mobile-friendly lightweight convolutions to form hybrid architectures. However, ViT and its variants still have higher latency or considerably more parameters than lightweight CNNs, even true for the years-old MobileNet. In practice, latency and size are both crucial for efficient deployment on resource-constraint hardware. In this work, we investigate a central question, can transformer models run as fast as MobileNet and maintain a similar size? We revisit the design choices of ViTs and propose an improved supernet with low latency and high parameter efficiency. We further introduce a fine-grained joint search strategy that can find efficient architectures by optimizing latency and number of parameters simultaneously. The proposed models, EfficientFormerV2, achieve about $4\%$ higher top-1 accuracy than MobileNetV2 and MobileNetV2$\times1.4$ on ImageNet-1K with similar latency and parameters. We demonstrate that properly designed and optimized vision transformers can achieve high performance with MobileNet-level size and speed.

视觉变压器（VIT）显示了计算机视觉任务的快速进步，在各种基准上取得了令人鼓舞的结果。但是，由于参数和模型设计的数量大量，例如注意机制，基于VIT的模型通常比轻型卷积网络慢。因此，为实时应用程序部署VIT特别具有挑战性，尤其是在资源受限的硬件（例如移动设备）上。最近的努力试图通过网络体系结构搜索或与Mobilenet块的混合设计来降低VIT的计算复杂性，但推理速度仍然不令人满意。这导致了一个重要的问题：变形金刚在获得高性能的同时可以像Mobilenet一样快吗？为了回答这一点，我们首先重新审视基于VIT的模型中使用的网络体系结构和运营商，并确定效率低下的设计。然后，我们引入了一个尺寸一致的纯变压器（无需Mobilenet块）作为设计范式。最后，我们执行以延迟驱动的缩小，以获取一系列称为EfficityFormer的最终模型。广泛的实验表明，在移动设备上的性能和速度方面，有效形式的优势。我们最快的型号，EfficientFormer-L1，在ImagEnet-1k上获得$ 79.2 \％$ $ TOP-1的准确性，仅$ 1.6 $ MS推理潜伏期在iPhone 12上（与Coreml一起编译），该{运行速度与MobileNetV2 $ \ Times Times 1.4 $（ $ 1.6 $ MS，$ 74.7 \％$ top-1），我们最大的型号EfficientFormer-L7，获得了$ 83.3 \％$精度，仅$ 7.0 $ MS延迟。我们的工作证明，正确设计的变压器可以在移动设备上达到极低的延迟，同时保持高性能。

View-dependent effects such as reflections pose a substantial challenge for image-based and neural rendering algorithms. Above all, curved reflectors are particularly hard, as they lead to highly non-linear reflection flows as the camera moves. We introduce a new point-based representation to compute Neural Point Catacaustics allowing novel-view synthesis of scenes with curved reflectors, from a set of casually-captured input photos. At the core of our method is a neural warp field that models catacaustic trajectories of reflections, so complex specular effects can be rendered using efficient point splatting in conjunction with a neural renderer. One of our key contributions is the explicit representation of reflections with a reflection point cloud which is displaced by the neural warp field, and a primary point cloud which is optimized to represent the rest of the scene. After a short manual annotation step, our approach allows interactive high-quality renderings of novel views with accurate reflection flow. Additionally, the explicit representation of reflection flow supports several forms of scene manipulation in captured scenes, such as reflection editing, cloning of specular objects, reflection tracking across views, and comfortable stereo viewing. We provide the source code and other supplemental material on https://repo-sam.inria.fr/ fungraph/neural_catacaustics/

Modern speech recognition systems exhibits rapid performance degradation under domain shift. This issue is especially prevalent in data-scarce settings, such as low-resource languages, where diversity of training data is limited. In this work we propose M2DS2, a simple and sample-efficient finetuning strategy for large pretrained speech models, based on mixed source and target domain self-supervision. We find that including source domain self-supervision stabilizes training and avoids mode collapse of the latent representations. For evaluation, we collect HParl, a $120$ hour speech corpus for Greek, consisting of plenary sessions in the Greek Parliament. We merge HParl with two popular Greek corpora to create GREC-MD, a test-bed for multi-domain evaluation of Greek ASR systems. In our experiments we find that, while other Unsupervised Domain Adaptation baselines fail in this resource-constrained environment, M2DS2 yields significant improvements for cross-domain adaptation, even when a only a few hours of in-domain audio are available. When we relax the problem in a weakly supervised setting, we find that independent adaptation for audio using M2DS2 and language using simple LM augmentation techniques is particularly effective, yielding word error rates comparable to the fully supervised baselines.

In this work, we propose a novel framework for estimating the dimension of the data manifold using a trained diffusion model. A trained diffusion model approximates the gradient of the log density of a noise-corrupted version of the target distribution for varying levels of corruption. If the data concentrates around a manifold embedded in the high-dimensional ambient space, then as the level of corruption decreases, the score function points towards the manifold, as this direction becomes the direction of maximum likelihood increase. Therefore, for small levels of corruption, the diffusion model provides us with access to an approximation of the normal bundle of the data manifold. This allows us to estimate the dimension of the tangent space, thus, the intrinsic dimension of the data manifold. Our method outperforms linear methods for dimensionality detection such as PPCA in controlled experiments.

This project leverages advances in multi-agent reinforcement learning (MARL) to improve the efficiency and flexibility of order-picking systems for commercial warehouses. We envision a warehouse of the future in which dozens of mobile robots and human pickers work together to collect and deliver items within the warehouse. The fundamental problem we tackle, called the order-picking problem, is how these worker agents must coordinate their movement and actions in the warehouse to maximise performance (e.g. order throughput) under given resource constraints. Established industry methods using heuristic approaches require large engineering efforts to optimise for innately variable warehouse configurations. In contrast, the MARL framework can be flexibly applied to any warehouse configuration (e.g. size, layout, number/types of workers, item replenishment frequency) and the agents learn via a process of trial-and-error how to optimally cooperate with one another. This paper details the current status of the R&D effort initiated by Dematic and the University of Edinburgh towards a general-purpose and scalable MARL solution for the order-picking problem in realistic warehouses.

Multimodal learning pipelines have benefited from the success of pretrained language models. However, this comes at the cost of increased model parameters. In this work, we propose Adapted Multimodal BERT (AMB), a BERT-based architecture for multimodal tasks that uses a combination of adapter modules and intermediate fusion layers. The adapter adjusts the pretrained language model for the task at hand, while the fusion layers perform task-specific, layer-wise fusion of audio-visual information with textual BERT representations. During the adaptation process the pre-trained language model parameters remain frozen, allowing for fast, parameter-efficient training. In our ablations we see that this approach leads to efficient models, that can outperform their fine-tuned counterparts and are robust to input noise. Our experiments on sentiment analysis with CMU-MOSEI show that AMB outperforms the current state-of-the-art across metrics, with 3.4% relative reduction in the resulting error and 2.1% relative improvement in 7-class classification accuracy.

Holistic methods using CNNs and margin-based losses have dominated research on face recognition. In this work, we depart from this setting in two ways: (a) we employ the Vision Transformer as an architecture for training a very strong baseline for face recognition, simply called fViT, which already surpasses most state-of-the-art face recognition methods. (b) Secondly, we capitalize on the Transformer's inherent property to process information (visual tokens) extracted from irregular grids to devise a pipeline for face recognition which is reminiscent of part-based face recognition methods. Our pipeline, called part fViT, simply comprises a lightweight network to predict the coordinates of facial landmarks followed by the Vision Transformer operating on patches extracted from the predicted landmarks, and it is trained end-to-end with no landmark supervision. By learning to extract discriminative patches, our part-based Transformer further boosts the accuracy of our Vision Transformer baseline achieving state-of-the-art accuracy on several face recognition benchmarks.

In contrast to the rapid digitalization of several industries, agriculture suffers from low adoption of smart farming tools. While AI-driven digital agriculture tools can offer high-performing predictive functionalities, they lack tangible quantitative evidence on their benefits to the farmers. Field experiments can derive such evidence, but are often costly, time consuming and hence limited in scope and scale of application. To this end, we propose an observational causal inference framework for the empirical evaluation of the impact of digital tools on target farm performance indicators (e.g., yield in this case). This way, we can increase farmers' trust via enhancing the transparency of the digital agriculture market and accelerate the adoption of technologies that aim to secure farmer income resilience and global agricultural sustainability. As a case study, we designed and implemented a recommendation system for the optimal sowing time of cotton based on numerical weather predictions, which was used by a farmers' cooperative during the growing season of 2021. We then leverage agricultural knowledge, collected yield data, and environmental information to develop a causal graph of the farm system. Using the back-door criterion, we identify the impact of sowing recommendations on the yield and subsequently estimate it using linear regression, matching, inverse propensity score weighting and meta-learners. The results reveal that a field sown according to our recommendations exhibited a statistically significant yield increase that ranged from 12% to 17%, depending on the method. The effect estimates were robust, as indicated by the agreement among the estimation methods and four successful refutation tests. We argue that this approach can be implemented for decision support systems of other fields, extending their evaluation beyond a performance assessment of internal functionalities.

Distributed deep learning (DDL) systems strongly depend on network performance. Current electronic packet switched (EPS) network architectures and technologies suffer from variable diameter topologies, low-bisection bandwidth and over-subscription affecting completion time of communication and collective operations. We introduce a near-exascale, full-bisection bandwidth, all-to-all, single-hop, all-optical network architecture with nanosecond reconfiguration called RAMP, which supports large-scale distributed and parallel computing systems (12.8~Tbps per node for up to 65,536 nodes). For the first time, a custom RAMP-x MPI strategy and a network transcoder is proposed to run MPI collective operations across the optical circuit switched (OCS) network in a schedule-less and contention-less manner. RAMP achieves 7.6-171$\times$ speed-up in completion time across all MPI operations compared to realistic EPS and OCS counterparts. It can also deliver a 1.3-16$\times$ and 7.8-58$\times$ reduction in Megatron and DLRM training time respectively} while offering 42-53$\times$ and 3.3-12.4$\times$ improvement in energy consumption and cost respectively.