签名的网络使我们能够对双方的关系和互动进行建模，例如朋友/敌人，支持/反对等。这些交互通常在真实数据集中是暂时的，在这些数据集中，节点和边缘会随时间出现。因此，学习签名网络的动态对于有效预测未来联系的符号和强度至关重要。现有的作品模型签名网络或动态网络，但并非都在一起。在这项工作中，我们研究了动态签名的网络，在这些网络中，链接都随时间签名和演变。我们的模型使用内存模块和平衡聚合（因此，名称SEMBA）学习了签名的链接的演变。每个节点都维护两个单独的内存编码，以实现正相互作用和负相互作用。在新边缘的到来时，每个交互节点汇总了此签名的信息，并利用平衡理论。节点嵌入是使用更新的内存生成的，然后将其用于训练多个下游任务，包括链接标志预测和链接权重预测。我们的结果表明，SEMBA的表现优于所有基准，即通过获得AUC增长8％，而FPR降低了50％。关于预测签名权重的任务的结果表明，SEMBA将平方误差降低了9％，同时降低了KL-Divergence对预测签名权重的分布的减少69％。

Prompting large language models has enabled significant recent progress in multi-step reasoning over text. However, when applied to text generation from semi-structured data (e.g., graphs or tables), these methods typically suffer from low semantic coverage, hallucination, and logical inconsistency. We propose MURMUR, a neuro-symbolic modular approach to text generation from semi-structured data with multi-step reasoning. MURMUR is a best-first search method that generates reasoning paths using: (1) neural and symbolic modules with specific linguistic and logical skills, (2) a grammar whose production rules define valid compositions of modules, and (3) value functions that assess the quality of each reasoning step. We conduct experiments on two diverse data-to-text generation tasks like WebNLG and LogicNLG. These tasks differ in their data representations (graphs and tables) and span multiple linguistic and logical skills. MURMUR obtains significant improvements over recent few-shot baselines like direct prompting and chain-of-thought prompting, while also achieving comparable performance to fine-tuned GPT-2 on out-of-domain data. Moreover, human evaluation shows that MURMUR generates highly faithful and correct reasoning paths that lead to 26% more logically consistent summaries on LogicNLG, compared to direct prompting.

Vision transformers (ViTs) have achieved impressive results on various computer vision tasks in the last several years. In this work, we study the capability of frozen ViTs, pretrained only on visual data, to generalize to audio-visual data without finetuning any of its original parameters. To do so, we propose a latent audio-visual hybrid (LAVISH) adapter that adapts pretrained ViTs to audio-visual tasks by injecting a small number of trainable parameters into every layer of a frozen ViT. To efficiently fuse visual and audio cues, our LAVISH adapter uses a small set of latent tokens, which form an attention bottleneck, thus, eliminating the quadratic cost of standard cross-attention. Compared to the existing modality-specific audio-visual methods, our approach achieves competitive or even better performance on various audio-visual tasks while using fewer tunable parameters and without relying on costly audio pretraining or external audio encoders. Our code is available at https://genjib.github.io/project_page/LAVISH/

The last several years have witnessed remarkable progress in video-and-language (VidL) understanding. However, most modern VidL approaches use complex and specialized model architectures and sophisticated pretraining protocols, making the reproducibility, analysis and comparisons of these frameworks difficult. Hence, instead of proposing yet another new VidL model, this paper conducts a thorough empirical study demystifying the most important factors in the VidL model design. Among the factors that we investigate are (i) the spatiotemporal architecture design, (ii) the multimodal fusion schemes, (iii) the pretraining objectives, (iv) the choice of pretraining data, (v) pretraining and finetuning protocols, and (vi) dataset and model scaling. Our empirical study reveals that the most important design factors include: temporal modeling, video-to-text multimodal fusion, masked modeling objectives, and joint training on images and videos. Using these empirical insights, we then develop a step-by-step recipe, dubbed VindLU, for effective VidL pretraining. Our final model trained using our recipe achieves comparable or better than state-of-the-art results on several VidL tasks without relying on external CLIP pretraining. In particular, on the text-to-video retrieval task, our approach obtains 61.2% on DiDeMo, and 55.0% on ActivityNet, outperforming current SOTA by 7.8% and 6.1% respectively. Furthermore, our model also obtains state-of-the-art video question-answering results on ActivityNet-QA, MSRVTT-QA, MSRVTT-MC and TVQA. Our code and pretrained models are publicly available at: https://github.com/klauscc/VindLU.

Fingerphoto images captured using a smartphone are successfully used to verify the individuals that have enabled several applications. This work presents a novel algorithm for fingerphoto verification using a nested residual block: Finger-NestNet. The proposed Finger-NestNet architecture is designed with three consecutive convolution blocks followed by a series of nested residual blocks to achieve reliable fingerphoto verification. This paper also presents the interpretability of the proposed method using four different visualization techniques that can shed light on the critical regions in the fingerphoto biometrics that can contribute to the reliable verification performance of the proposed method. Extensive experiments are performed on the fingerphoto dataset comprised of 196 unique fingers collected from 52 unique data subjects using an iPhone6S. Experimental results indicate the improved verification of the proposed method compared to six different existing methods with EER = 1.15%.

We propose Universal Document Processing (UDOP), a foundation Document AI model which unifies text, image, and layout modalities together with varied task formats, including document understanding and generation. UDOP leverages the spatial correlation between textual content and document image to model image, text, and layout modalities with one uniform representation. With a novel Vision-Text-Layout Transformer, UDOP unifies pretraining and multi-domain downstream tasks into a prompt-based sequence generation scheme. UDOP is pretrained on both large-scale unlabeled document corpora using innovative self-supervised objectives and diverse labeled data. UDOP also learns to generate document images from text and layout modalities via masked image reconstruction. To the best of our knowledge, this is the first time in the field of document AI that one model simultaneously achieves high-quality neural document editing and content customization. Our method sets the state-of-the-art on 9 Document AI tasks, e.g., document understanding and QA, across diverse data domains like finance reports, academic papers, and websites. UDOP ranks first on the leaderboard of the Document Understanding Benchmark (DUE).

Recent datasets expose the lack of the systematic generalization ability in standard sequence-to-sequence models. In this work, we analyze this behavior of seq2seq models and identify two contributing factors: a lack of mutual exclusivity bias (i.e., a source sequence already mapped to a target sequence is less likely to be mapped to other target sequences), and the tendency to memorize whole examples rather than separating structures from contents. We propose two techniques to address these two issues respectively: Mutual Exclusivity Training that prevents the model from producing seen generations when facing novel, unseen examples via an unlikelihood-based loss; and prim2primX data augmentation that automatically diversifies the arguments of every syntactic function to prevent memorizing and provide a compositional inductive bias without exposing test-set data. Combining these two techniques, we show substantial empirical improvements using standard sequence-to-sequence models (LSTMs and Transformers) on two widely-used compositionality datasets: SCAN and COGS. Finally, we provide analysis characterizing the improvements as well as the remaining challenges, and provide detailed ablations of our method. Our code is available at https://github.com/owenzx/met-primaug

This paper addresses the problem of position estimation in UAVs operating in a cluttered environment where GPS information is unavailable. A model learning-based approach is proposed that takes in the rotor RPMs and past state as input and predicts the one-step-ahead position of the UAV using a novel spectral-normalized memory neural network (SN-MNN). The spectral normalization guarantees stable and reliable prediction performance. The predicted position is transformed to global coordinate frame which is then fused along with the odometry of other peripheral sensors like IMU, barometer, compass etc., using the onboard extended Kalman filter to estimate the states of the UAV. The experimental flight data collected from a motion capture facility using a micro-UAV is used to train the SN-MNN. The PX4-ECL library is used to replay the flight data using the proposed algorithm, and the estimated position is compared with actual ground truth data. The proposed algorithm doesn't require any additional onboard sensors, and is computationally light. The performance of the proposed approach is compared with the current state-of-art GPS-denied algorithms, and it can be seen that the proposed algorithm has the least RMSE for position estimates.

Humans exhibit disagreement during data labeling. We term this disagreement as human label uncertainty. In this work, we study the ramifications of human label uncertainty (HLU). Our evaluation of existing uncertainty estimation algorithms, with the presence of HLU, indicates the limitations of existing uncertainty metrics and algorithms themselves in response to HLU. Meanwhile, we observe undue effects in predictive uncertainty and generalizability. To mitigate the undue effects, we introduce a novel natural scene statistics (NSS) based label dilution training scheme without requiring massive human labels. Specifically, we first select a subset of samples with low perceptual quality ranked by statistical regularities of images. We then assign separate labels to each sample in this subset to obtain a training set with diluted labels. Our experiments and analysis demonstrate that training with NSS-based label dilution alleviates the undue effects caused by HLU.

Time-resolved image sensors that capture light at pico-to-nanosecond timescales were once limited to niche applications but are now rapidly becoming mainstream in consumer devices. We propose low-cost and low-power imaging modalities that capture scene information from minimal time-resolved image sensors with as few as one pixel. The key idea is to flood illuminate large scene patches (or the entire scene) with a pulsed light source and measure the time-resolved reflected light by integrating over the entire illuminated area. The one-dimensional measured temporal waveform, called \emph{transient}, encodes both distances and albedoes at all visible scene points and as such is an aggregate proxy for the scene's 3D geometry. We explore the viability and limitations of the transient waveforms by themselves for recovering scene information, and also when combined with traditional RGB cameras. We show that plane estimation can be performed from a single transient and that using only a few more it is possible to recover a depth map of the whole scene. We also show two proof-of-concept hardware prototypes that demonstrate the feasibility of our approach for compact, mobile, and budget-limited applications.