Unsupervised video object learning seeks to decompose video scenes into structural object representations without any supervision from depth, optical flow, or segmentation. We present VONet, an innovative approach that is inspired by MONet. While utilizing a U-Net architecture, VONet employs an efficient and effective parallel attention inference process, generating attention masks for all slots simultaneously. Additionally, to enhance the temporal consistency of each mask across consecutive video frames, VONet develops an object-wise sequential VAE framework. The integration of these innovative encoder-side techniques, in conjunction with an expressive transformer-based decoder, establishes VONet as the leading unsupervised method for object learning across five MOVI datasets, encompassing videos of diverse complexities. Code is available at https://github.com/hnyu/vonet.
In this research, we introduce the enhanced automated quality assessment network (IBS-AQSNet), an innovative solution for assessing the quality of interactive building segmentation within high-resolution remote sensing imagery. This is a new challenge in segmentation quality assessment, and our proposed IBS-AQSNet allievate this by identifying missed and mistaken segment areas. First of all, to acquire robust image features, our method combines a robust, pre-trained backbone with a lightweight counterpart for comprehensive feature extraction from imagery and segmentation results. These features are then fused through a simple combination of concatenation, convolution layers, and residual connections. Additionally, ISR-AQSNet incorporates a multi-scale differential quality assessment decoder, proficient in pinpointing areas where segmentation result is either missed or mistaken. Experiments on a newly-built EVLab-BGZ dataset, which includes over 39,198 buildings, demonstrate the superiority of the proposed method in automating segmentation quality assessment, thereby setting a new benchmark in the field.
Accurate liver segmentation from CT scans is essential for effective diagnosis and treatment planning. Computer-aided diagnosis systems promise to improve the precision of liver disease diagnosis, disease progression, and treatment planning. In response to the need, we propose a novel deep learning approach, \textit{\textbf{PVTFormer}}, that is built upon a pretrained pyramid vision transformer (PVT v2) combined with advanced residual upsampling and decoder block. By integrating a refined feature channel approach with a hierarchical decoding strategy, PVTFormer generates high quality segmentation masks by enhancing semantic features. Rigorous evaluation of the proposed method on Liver Tumor Segmentation Benchmark (LiTS) 2017 demonstrates that our proposed architecture not only achieves a high dice coefficient of 86.78\%, mIoU of 78.46\%, but also obtains a low HD of 3.50. The results underscore PVTFormer's efficacy in setting a new benchmark for state-of-the-art liver segmentation methods. The source code of the proposed PVTFormer is available at \url{https://github.com/DebeshJha/PVTFormer}.
A common pipeline in functional data analysis is to first convert the discretely observed data to smooth functions, and then represent the functions by a finite-dimensional vector of coefficients summarizing the information. Existing methods for data smoothing and dimensional reduction mainly focus on learning the linear mappings from the data space to the representation space, however, learning only the linear representations may not be sufficient. In this study, we propose to learn the nonlinear representations of functional data using neural network autoencoders designed to process data in the form it is usually collected without the need of preprocessing. We design the encoder to employ a projection layer computing the weighted inner product of the functional data and functional weights over the observed timestamp, and the decoder to apply a recovery layer that maps the finite-dimensional vector extracted from the functional data back to functional space using a set of predetermined basis functions. The developed architecture can accommodate both regularly and irregularly spaced data. Our experiments demonstrate that the proposed method outperforms functional principal component analysis in terms of prediction and classification, and maintains superior smoothing ability and better computational efficiency in comparison to the conventional autoencoders under both linear and nonlinear settings.
Cloud radiative feedback impacts early tropical cyclone (TC) intensification, but limitations in existing diagnostic frameworks make them unsuitable for studying asymmetric or transient radiative heating. We propose a linear Variational Encoder-Decoder (VED) to learn the hidden relationship between radiation and the surface intensification of realistic simulated TCs. Limiting VED model inputs enables using its uncertainty to identify periods when radiation has more importance for intensification. A close examination of the extracted 3D radiative structures suggests that longwave radiative forcing from inner core deep convection and shallow clouds both contribute to intensification, with the deep convection having the most impact overall. We find that deep convection downwind of the shallow clouds is critical to the intensification of Haiyan. Our work demonstrates that machine learning can discover thermodynamic-kinematic relationships without relying on axisymmetric or deterministic assumptions, paving the way towards the objective discovery of processes leading to TC intensification in realistic conditions.
In this work, we address various segmentation tasks, each traditionally tackled by distinct or partially unified models. We propose OMG-Seg, One Model that is Good enough to efficiently and effectively handle all the segmentation tasks, including image semantic, instance, and panoptic segmentation, as well as their video counterparts, open vocabulary settings, prompt-driven, interactive segmentation like SAM, and video object segmentation. To our knowledge, this is the first model to handle all these tasks in one model and achieve satisfactory performance. We show that OMG-Seg, a transformer-based encoder-decoder architecture with task-specific queries and outputs, can support over ten distinct segmentation tasks and yet significantly reduce computational and parameter overhead across various tasks and datasets. We rigorously evaluate the inter-task influences and correlations during co-training. Code and models are available at https://github.com/lxtGH/OMG-Seg.
Advanced by transformer architecture, vision foundation models (VFMs) achieve remarkable progress in performance and generalization ability. Segment Anything Model (SAM) is one remarkable model that can achieve generalized segmentation. However, most VFMs cannot run in realtime, which makes it difficult to transfer them into several products. On the other hand, current real-time segmentation mainly has one purpose, such as semantic segmentation on the driving scene. We argue that diverse outputs are needed for real applications. Thus, this work explores a new real-time segmentation setting, named all-purpose segmentation in real-time, to transfer VFMs in real-time deployment. It contains three different tasks, including interactive segmentation, panoptic segmentation, and video segmentation. We aim to use one model to achieve the above tasks in real-time. We first benchmark several strong baselines. Then, we present Real-Time All Purpose SAM (RAP-SAM). It contains an efficient encoder and an efficient decoupled decoder to perform prompt-driven decoding. Moreover, we further explore different training strategies and tuning methods to boost co-training performance further. Our code and model are available at https://github.com/xushilin1/RAP-SAM/.
Scene Text Recognition (STR) is an important and challenging upstream task for building structured information databases, that involves recognizing text within images of natural scenes. Although current state-of-the-art (SOTA) models for STR exhibit high performance, they typically suffer from low inference efficiency due to their reliance on hybrid architectures comprised of visual encoders and sequence decoders. In this work, we propose a VIsion Permutable extractor for fast and efficient Scene Text Recognition (SVIPTR), which achieves an impressive balance between high performance and rapid inference speeds in the domain of STR. Specifically, SVIPTR leverages a visual-semantic extractor with a pyramid structure, characterized by the Permutation and combination of local and global self-attention layers. This design results in a lightweight and efficient model and its inference is insensitive to input length. Extensive experimental results on various standard datasets for both Chinese and English scene text recognition validate the superiority of SVIPTR. Notably, the SVIPTR-T (Tiny) variant delivers highly competitive accuracy on par with other lightweight models and achieves SOTA inference speeds. Meanwhile, the SVIPTR-L (Large) attains SOTA accuracy in single-encoder-type models, while maintaining a low parameter count and favorable inference speed. Our proposed method provides a compelling solution for the STR challenge, which greatly benefits real-world applications requiring fast and efficient STR. The code is publicly available at https://github.com/cxfyxl/VIPTR.
Predictive coding-inspired deep networks for visual computing integrate classification and reconstruction processes in shared intermediate layers. Although synergy between these processes is commonly assumed, it has yet to be convincingly demonstrated. In this study, we take a critical look at how classifying and reconstructing interact in deep learning architectures. Our approach utilizes a purposefully designed family of model architectures reminiscent of autoencoders, each equipped with an encoder, a decoder, and a classification head featuring varying modules and complexities. We meticulously analyze the extent to which classification- and reconstruction-driven information can seamlessly coexist within the shared latent layer of the model architectures. Our findings underscore a significant challenge: Classification-driven information diminishes reconstruction-driven information in intermediate layers' shared representations and vice versa. While expanding the shared representation's dimensions or increasing the network's complexity can alleviate this trade-off effect, our results challenge prevailing assumptions in predictive coding and offer guidance for future iterations of predictive coding concepts in deep networks.
Precisely determining the contact force during safe interaction in Minimally Invasive Robotic Surgery (MIRS) is still an open research challenge. Inspired by post-operative qualitative analysis from surgical videos, the use of cross-modality data driven deep neural network models has been one of the newest approaches to predict sensorless force trends. However, these methods required for large and variable datasets which are not currently available. In this paper, we present a new vision-haptic dataset (DaFoEs) with variable soft environments for the training of deep neural models. In order to reduce the bias from a single dataset, we present a pipeline to generalize different vision and state data inputs for mixed dataset training, using a previously validated dataset with different setup. Finally, we present a variable encoder-decoder architecture to predict the forces done by the laparoscopic tool using single input or sequence of inputs. For input sequence, we use a recurrent decoder, named with the prefix R, and a new temporal sampling to represent the acceleration of the tool. During our training, we demonstrate that single dataset training tends to overfit to the training data domain, but has difficulties on translating the results across new domains. However, dataset mixing presents a good translation with a mean relative estimated force error of 5% and 12% for the recurrent and non-recurrent models respectively. Our method, also marginally increase the effectiveness of transformers for force estimation up to a maximum of ~15%, as the volume of available data is increase by 150%. In conclusion, we demonstrate that mixing experimental set ups for vision-state force estimation in MIRS is a possible approach towards the general solution of the problem.
Moderate-sized large language models (LLMs) -- those with 7B or 13B parameters -- exhibit promising machine translation (MT) performance. However, even the top-performing 13B LLM-based translation models, like ALMA, does not match the performance of state-of-the-art conventional encoder-decoder translation models or larger-scale LLMs such as GPT-4. In this study, we bridge this performance gap. We first assess the shortcomings of supervised fine-tuning for LLMs in the MT task, emphasizing the quality issues present in the reference data, despite being human-generated. Then, in contrast to SFT which mimics reference translations, we introduce Contrastive Preference Optimization (CPO), a novel approach that trains models to avoid generating adequate but not perfect translations. Applying CPO to ALMA models with only 22K parallel sentences and 12M parameters yields significant improvements. The resulting model, called ALMA-R, can match or exceed the performance of the WMT competition winners and GPT-4 on WMT'21, WMT'22 and WMT'23 test datasets.
This paper delineates the visual speech recognition (VSR) system introduced by the NPU-ASLP-LiAuto (Team 237) in the first Chinese Continuous Visual Speech Recognition Challenge (CNVSRC) 2023, engaging in the fixed and open tracks of Single-Speaker VSR Task, and the open track of Multi-Speaker VSR Task. In terms of data processing, we leverage the lip motion extractor from the baseline1 to produce multi-scale video data. Besides, various augmentation techniques are applied during training, encompassing speed perturbation, random rotation, horizontal flipping, and color transformation. The VSR model adopts an end-to-end architecture with joint CTC/attention loss, comprising a ResNet3D visual frontend, an E-Branchformer encoder, and a Transformer decoder. Experiments show that our system achieves 34.76% CER for the Single-Speaker Task and 41.06% CER for the Multi-Speaker Task after multi-system fusion, ranking first place in all three tracks we participate.
Pretraining monolingual language models have been proven to be vital for performance in Arabic Natural Language Processing (NLP) tasks. In this paper, we conduct a comprehensive study on the role of data in Arabic Pretrained Language Models (PLMs). More precisely, we reassess the performance of a suite of state-of-the-art Arabic PLMs by retraining them on massive-scale, high-quality Arabic corpora. We have significantly improved the performance of the leading Arabic encoder-only BERT-base and encoder-decoder T5-base models on the ALUE and ORCA leaderboards, thereby reporting state-of-the-art results in their respective model categories. In addition, our analysis strongly suggests that pretraining data by far is the primary contributor to performance, surpassing other factors. Our models and source code are publicly available at https://github.com/huawei-noah/Pretrained-Language-Model/tree/master/JABER-PyTorch.
Masked graph modeling excels in the self-supervised representation learning of molecular graphs. Scrutinizing previous studies, we can reveal a common scheme consisting of three key components: (1) graph tokenizer, which breaks a molecular graph into smaller fragments (i.e., subgraphs) and converts them into tokens; (2) graph masking, which corrupts the graph with masks; (3) graph autoencoder, which first applies an encoder on the masked graph to generate the representations, and then employs a decoder on the representations to recover the tokens of the original graph. However, the previous MGM studies focus extensively on graph masking and encoder, while there is limited understanding of tokenizer and decoder. To bridge the gap, we first summarize popular molecule tokenizers at the granularity of node, edge, motif, and Graph Neural Networks (GNNs), and then examine their roles as the MGM's reconstruction targets. Further, we explore the potential of adopting an expressive decoder in MGM. Our results show that a subgraph-level tokenizer and a sufficiently expressive decoder with remask decoding have a large impact on the encoder's representation learning. Finally, we propose a novel MGM method SimSGT, featuring a Simple GNN-based Tokenizer (SGT) and an effective decoding strategy. We empirically validate that our method outperforms the existing molecule self-supervised learning methods. Our codes and checkpoints are available at https://github.com/syr-cn/SimSGT.
Unlike single image task, stereo image enhancement can use another view information, and its key stage is how to perform cross-view feature interaction to extract useful information from another view. However, complex noise in low-light image and its impact on subsequent feature encoding and interaction are ignored by the existing methods. In this paper, a method is proposed to perform enhancement and de-noising simultaneously. First, to reduce unwanted noise interference, a low-frequency information enhanced module (IEM) is proposed to suppress noise and produce a new image space. Additionally, a cross-channel and spatial context information mining module (CSM) is proposed to encode long-range spatial dependencies and to enhance inter-channel feature interaction. Relying on CSM, an encoder-decoder structure is constructed, incorporating cross-view and cross-scale feature interactions to perform enhancement in the new image space. Finally, the network is trained with the constraints of both spatial and frequency domain losses. Extensive experiments on both synthesized and real datasets show that our method obtains better detail recovery and noise removal compared with state-of-the-art methods. In addition, a real stereo image enhancement dataset is captured with stereo camera ZED2. The code and dataset are publicly available at: https://www.github.com/noportraits/LFENet.
Addition is perhaps one of the simplest arithmetic tasks one can think of and is usually performed using the carrying over algorithm. This algorithm consists of two tasks: adding digits in the same position and carrying over a one whenever necessary. We study how transformer models implement this algorithm and how the two aforementioned tasks are allocated to different parts of the network. We first focus on two-layer encoder-only models and show that the carrying over algorithm is implemented in a modular fashion. The first layer is mostly responsible for adding digits in the same position. The second layer first decides, in the attention, which positions need a carried one or not, and then performs the carrying of the one in the final MLP. We provide a simple way of precisely identifying which neurons are responsible for that task. This implementation of the carrying over algorithm occurs across a range of hyperparameters for two as well as three-layer models. For small decoder-only models, we observe the same implementation and provide suggestive evidence for its existence in three 7B large language models.
The ever-increasing volume of paper submissions makes it difficult to stay informed about the latest state-of-the-art research. To address this challenge, we introduce LEGOBench, a benchmark for evaluating systems that generate scientific leaderboards. LEGOBench is curated from 22 years of preprint submission data on arXiv and more than 11k machine learning leaderboards on the PapersWithCode portal. We present four graph-based and two language model-based leaderboard generation task configurations. We evaluate popular encoder-only scientific language models as well as decoder-only large language models across these task configurations. State-of-the-art models showcase significant performance gaps in automatic leaderboard generation on LEGOBench. The code is available on GitHub ( https://github.com/lingo-iitgn/LEGOBench ) and the dataset is hosted on OSF ( https://osf.io/9v2py/?view_only=6f91b0b510df498ba01595f8f278f94c ).
Transformers are considered conceptually different from the previous generation of state-of-the-art NLP models - recurrent neural networks (RNNs). In this work, we demonstrate that decoder-only transformers can in fact be conceptualized as unbounded multi-state RNNs - an RNN variant with unlimited hidden state size. We further show that transformers can be converted into $\textit{bounded}$ multi-state RNNs by fixing the size of their hidden state, effectively compressing their key-value cache. We introduce a novel, training-free compression policy - $\textbf{T}$oken $\textbf{O}$mission $\textbf{V}$ia $\textbf{A}$ttention (TOVA). Our experiments with four long range tasks and several LLMs show that TOVA outperforms several baseline compression policies. Particularly, our results are nearly on par with the full model, using in some cases only $\frac{1}{8}$ of the original cache size, which translates to 4.8X higher throughput. Our results shed light on the connection between transformers and RNNs, and help mitigate one of LLMs' most painful computational bottlenecks - the size of their key-value cache. We publicly release our code at https://github.com/schwartz-lab-NLP/TOVA
Traffic forecasting is the foundation for intelligent transportation systems. Spatiotemporal graph neural networks have demonstrated state-of-the-art performance in traffic forecasting. However, these methods do not explicitly model some of the natural characteristics in traffic data, such as the multiscale structure that encompasses spatial and temporal variations at different levels of granularity or scale. To that end, we propose a Wavelet-Inspired Graph Convolutional Recurrent Network (WavGCRN) which combines multiscale analysis (MSA)-based method with Deep Learning (DL)-based method. In WavGCRN, the traffic data is decomposed into time-frequency components with Discrete Wavelet Transformation (DWT), constructing a multi-stream input structure; then Graph Convolutional Recurrent networks (GCRNs) are employed as encoders for each stream, extracting spatiotemporal features in different scales; and finally the learnable Inversed DWT and GCRN are combined as the decoder, fusing the information from all streams for traffic metrics reconstruction and prediction. Furthermore, road-network-informed graphs and data-driven graph learning are combined to accurately capture spatial correlation. The proposed method can offer well-defined interpretability, powerful learning capability, and competitive forecasting performance on real-world traffic data sets.
Counterfactual inference aims to answer retrospective "what if" questions and thus belongs to the most fine-grained type of inference in Pearl's causality ladder. Existing methods for counterfactual inference with continuous outcomes aim at point identification and thus make strong and unnatural assumptions about the underlying structural causal model. In this paper, we relax these assumptions and aim at partial counterfactual identification of continuous outcomes, i.e., when the counterfactual query resides in an ignorance interval with informative bounds. We prove that, in general, the ignorance interval of the counterfactual queries has non-informative bounds, already when functions of structural causal models are continuously differentiable. As a remedy, we propose a novel sensitivity model called Curvature Sensitivity Model. This allows us to obtain informative bounds by bounding the curvature of level sets of the functions. We further show that existing point counterfactual identification methods are special cases of our Curvature Sensitivity Model when the bound of the curvature is set to zero. We then propose an implementation of our Curvature Sensitivity Model in the form of a novel deep generative model, which we call Augmented Pseudo-Invertible Decoder. Our implementation employs (i) residual normalizing flows with (ii) variational augmentations. We empirically demonstrate the effectiveness of our Augmented Pseudo-Invertible Decoder. To the best of our knowledge, ours is the first partial identification model for Markovian structural causal models with continuous outcomes.