Creating editable videos that depict complex interactions between multiple objects in various artistic styles has long been a challenging task in filmmaking. Progress is often hampered by the scarcity of data sets that contain paired text descriptions and corresponding videos that showcase these interactions. This paper introduces a novel depth-conditioning approach that significantly advances this field by enabling the generation of coherent and diverse videos from just a single text-video pair using a pre-trained depth-aware Text-to-Image (T2I) model. Our method fine-tunes the pre-trained model to capture continuous motion by employing custom-designed spatial and temporal attention mechanisms. During inference, we use the DDIM inversion to provide structural guidance for video generation. This innovative technique allows for continuously controllable depth in videos, facilitating the generation of multiobject interactions while maintaining the concept generation and compositional strengths of the original T2I model across various artistic styles, such as photorealism, animation, and impressionism.
Achieving personalized intelligence at the edge with real-time learning capabilities holds enormous promise in enhancing our daily experiences and helping decision making, planning, and sensing. However, efficient and reliable edge learning remains difficult with current technology due to the lack of personalized data, insufficient hardware capabilities, and inherent challenges posed by online learning. Over time and across multiple developmental stages, the brain has evolved to efficiently incorporate new knowledge by gradually building on previous knowledge. In this work, we emulate the multiple stages of learning with digital neuromorphic technology that simulates the neural and synaptic processes of the brain using two stages of learning. First, a meta-training stage trains the hyperparameters of synaptic plasticity for one-shot learning using a differentiable simulation of the neuromorphic hardware. This meta-training process refines a hardware local three-factor synaptic plasticity rule and its associated hyperparameters to align with the trained task domain. In a subsequent deployment stage, these optimized hyperparameters enable fast, data-efficient, and accurate learning of new classes. We demonstrate our approach using event-driven vision sensor data and the Intel Loihi neuromorphic processor with its plasticity dynamics, achieving real-time one-shot learning of new classes that is vastly improved over transfer learning. Our methodology can be deployed with arbitrary plasticity models and can be applied to situations demanding quick learning and adaptation at the edge, such as navigating unfamiliar environments or learning unexpected categories of data through user engagement.
In recent years, Wi-Fi sensing has garnered significant attention due to its numerous benefits, such as privacy protection, low cost, and penetration ability. Extensive research has been conducted in this field, focusing on areas such as gesture recognition, people identification, and fall detection. However, many data-driven methods encounter challenges related to domain shift, where the model fails to perform well in environments different from the training data. One major factor contributing to this issue is the limited availability of Wi-Fi sensing datasets, which makes models learn excessive irrelevant information and over-fit to the training set. Unfortunately, collecting large-scale Wi-Fi sensing datasets across diverse scenarios is a challenging task. To address this problem, we propose CrossFi, a siamese network-based approach that excels in both in-domain scenario and cross-domain scenario, including few-shot, zero-shot scenarios, and even works in few-shot new-class scenario where testing set contains new categories. The core component of CrossFi is a sample-similarity calculation network called CSi-Net, which improves the structure of the siamese network by using an attention mechanism to capture similarity information, instead of simply calculating the distance or cosine similarity. Based on it, we develop an extra Weight-Net that can generate a template for each class, so that our CrossFi can work in different scenarios. Experimental results demonstrate that our CrossFi achieves state-of-the-art performance across various scenarios. In gesture recognition task, our CrossFi achieves an accuracy of 98.17% in in-domain scenario, 91.72% in one-shot cross-domain scenario, 64.81% in zero-shot cross-domain scenario, and 84.75% in one-shot new-class scenario. The code for our model is publicly available at https://github.com/RS2002/CrossFi.
Medical image segmentation is one of the domains where sufficient annotated data is not available. This necessitates the application of low-data frameworks like few-shot learning. Contemporary prototype-based frameworks often do not account for the variation in features within the support and query images, giving rise to a large variance in prototype alignment. In this work, we adopt a prototype-based self-supervised one-way one-shot learning framework using pseudo-labels generated from superpixels to learn the semantic segmentation task itself. We use a correlation-based probability score to generate a dynamic prototype for each query pixel from the bag of prototypes obtained from the support feature map. This weighting scheme helps to give a higher weightage to contextually related prototypes. We also propose a quadrant masking strategy in the downstream segmentation task by utilizing prior domain information to discard unwanted false positives. We present extensive experimentations and evaluations on abdominal CT and MR datasets to show that the proposed simple but potent framework performs at par with the state-of-the-art methods.
With the exponential growth of multimedia data, leveraging multimodal sensors presents a promising approach for improving accuracy in human activity recognition. Nevertheless, accurately identifying these activities using both video data and wearable sensor data presents challenges due to the labor-intensive data annotation, and reliance on external pretrained models or additional data. To address these challenges, we introduce Multimodal Masked Autoencoders-Based One-Shot Learning (Mu-MAE). Mu-MAE integrates a multimodal masked autoencoder with a synchronized masking strategy tailored for wearable sensors. This masking strategy compels the networks to capture more meaningful spatiotemporal features, which enables effective self-supervised pretraining without the need for external data. Furthermore, Mu-MAE leverages the representation extracted from multimodal masked autoencoders as prior information input to a cross-attention multimodal fusion layer. This fusion layer emphasizes spatiotemporal features requiring attention across different modalities while highlighting differences from other classes, aiding in the classification of various classes in metric-based one-shot learning. Comprehensive evaluations on MMAct one-shot classification show that Mu-MAE outperforms all the evaluated approaches, achieving up to an 80.17% accuracy for five-way one-shot multimodal classification, without the use of additional data.
Large machine-learning training datasets can be distilled into small collections of informative synthetic data samples. These synthetic sets support efficient model learning and reduce the communication cost of data sharing. Thus, high-fidelity distilled data can support the efficient deployment of machine learning applications in distributed network environments. A naive way to construct a synthetic set in a distributed environment is to allow each client to perform local data distillation and to merge local distillations at a central server. However, the quality of the resulting set is impaired by heterogeneity in the distributions of the local data held by clients. To overcome this challenge, we introduce the first collaborative data distillation technique, called CollabDM, which captures the global distribution of the data and requires only a single round of communication between client and server. Our method outperforms the state-of-the-art one-shot learning method on skewed data in distributed learning environments. We also show the promising practical benefits of our method when applied to attack detection in 5G networks.
Most existing deep learning-based registration methods are trained on single-type images to address same-domain tasks.However, cross-domain deformable registration remains challenging.We argue that the tailor-made matching criteria in traditional registration methods is one of the main reason they are applicable in different domains.Motivated by this, we devise a registration-oriented encoder to model the matching criteria of image features and structural features, which is beneficial to boost registration accuracy and adaptability.Specifically, a general feature encoder (Encoder-G) is proposed to capture comprehensive medical image features, while a structural feature encoder (Encoder-S) is designed to encode the structural self-similarity into the global representation.Extensive experiments on images from three different domains prove the efficacy of the proposed method. Moreover, by updating Encoder-S using one-shot learning, our method can effectively adapt to different domains.The code is publicly available at https://github.com/JuliusWang-7/EncoderReg.
Object detection is a critical component of transportation systems, particularly for applications such as autonomous driving, traffic monitoring, and infrastructure maintenance. Traditional object detection methods often struggle with limited data and variability in object appearance. The Oriented Window Learning Vision Transformer (OWL-ViT) offers a novel approach by adapting window orientations to the geometry and existence of objects, making it highly suitable for detecting diverse roadway assets. This study leverages OWL-ViT within a one-shot learning framework to recognize transportation infrastructure components, such as traffic signs, poles, pavement, and cracks. This study presents a novel method for roadway asset detection using OWL-ViT. We conducted a series of experiments to evaluate the performance of the model in terms of detection consistency, semantic flexibility, visual context adaptability, resolution robustness, and impact of non-max suppression. The results demonstrate the high efficiency and reliability of the OWL-ViT across various scenarios, underscoring its potential to enhance the safety and efficiency of intelligent transportation systems.
Deep learning (DL)-based Semantic Communications (SemCom) is becoming critical to maximize overall efficiency of communication networks. Nevertheless, SemCom is sensitive to wireless channel uncertainties, source outliers, and suffer from poor generalization bottlenecks. To address the mentioned challenges, this paper develops a latent diffusion model-enabled SemCom system with three key contributions, i.e., i) to handle potential outliers in the source data, semantic errors obtained by projected gradient descent based on the vulnerabilities of DL models, are utilized to update the parameters and obtain an outlier-robust encoder, ii) a lightweight single-layer latent space transformation adapter completes one-shot learning at the transmitter and is placed before the decoder at the receiver, enabling adaptation for out-of-distribution data and enhancing human-perceptual quality, and iii) an end-to-end consistency distillation (EECD) strategy is used to distill the diffusion models trained in latent space, enabling deterministic single or few-step low-latency denoising in various noisy channels while maintaining high semantic quality. Extensive numerical experiments across different datasets demonstrate the superiority of the proposed SemCom system, consistently proving its robustness to outliers, the capability to transmit data with unknown distributions, and the ability to perform real-time channel denoising tasks while preserving high human perceptual quality, outperforming the existing denoising approaches in semantic metrics like learned perceptual image path similarity (LPIPS).
The identification and restoration of ancient watermarks have long been a major topic in codicology and history. Classifying historical documents based on watermarks is challenging due to their diversity, noisy samples, multiple representation modes, and minor distinctions between classes and intra-class variations. This paper proposes a modified U-net-based conditional generative adversarial network (GAN) named Npix2Cpix to translate noisy raw historical watermarked images into clean, handwriting-free watermarked images by performing image translation from degraded (noisy) pixels to clean pixels. Using image-to-image translation and adversarial learning, the network creates clutter-free images for watermark restoration and categorization. The generator and discriminator of the proposed GAN are trained using two separate loss functions, each based on the distance between images, to learn the mapping from the input noisy image to the output clean image. After using the proposed GAN to pre-process noisy watermarked images, Siamese-based one-shot learning is employed for watermark classification. Experimental results on a large-scale historical watermark dataset demonstrate that cleaning the noisy watermarked images can help to achieve high one-shot classification accuracy. The qualitative and quantitative evaluation of the retrieved watermarked image highlights the effectiveness of the proposed approach.
Understanding how sentences are internally represented in the human brain, as well as in large language models (LLMs) such as ChatGPT, is a major challenge for cognitive science. Classic linguistic theories propose that the brain represents a sentence by parsing it into hierarchically organized constituents. In contrast, LLMs do not explicitly parse linguistic constituents and their latent representations remains poorly explained. Here, we demonstrate that humans and LLMs construct similar latent representations of hierarchical linguistic constituents by analyzing their behaviors during a novel one-shot learning task, in which they infer which words should be deleted from a sentence. Both humans and LLMs tend to delete a constituent, instead of a nonconstituent word string. In contrast, a naive sequence processing model that has access to word properties and ordinal positions does not show this property. Based on the word deletion behaviors, we can reconstruct the latent constituency tree representation of a sentence for both humans and LLMs. These results demonstrate that a latent tree-structured constituency representation can emerge in both the human brain and LLMs.
Machine learning has recently emerged as a powerful tool for generating new molecular and material structures. The success of state-of-the-art models stems from their ability to incorporate physical symmetries, such as translation, rotation, and periodicity. Here, we present a novel generative method called Response Matching (RM), which leverages the fact that each stable material or molecule exists at the minimum of its potential energy surface. Consequently, any perturbation induces a response in energy and stress, driving the structure back to equilibrium. Matching to such response is closely related to score matching in diffusion models. By employing the combination of a machine learning interatomic potential and random structure search as the denoising model, RM exploits the locality of atomic interactions, and inherently respects permutation, translation, rotation, and periodic invariances. RM is the first model to handle both molecules and bulk materials under the same framework. We demonstrate the efficiency and generalization of RM across three systems: a small organic molecular dataset, stable crystals from the Materials Project, and one-shot learning on a single diamond configuration.
Substance use disorders (SUDs) are a growing concern globally, necessitating enhanced understanding of the problem and its trends through data-driven research. Social media are unique and important sources of information about SUDs, particularly since the data in such sources are often generated by people with lived experiences. In this paper, we introduce Reddit-Impacts, a challenging Named Entity Recognition (NER) dataset curated from subreddits dedicated to discussions on prescription and illicit opioids, as well as medications for opioid use disorder. The dataset specifically concentrates on the lesser-studied, yet critically important, aspects of substance use--its clinical and social impacts. We collected data from chosen subreddits using the publicly available Application Programming Interface for Reddit. We manually annotated text spans representing clinical and social impacts reported by people who also reported personal nonmedical use of substances including but not limited to opioids, stimulants and benzodiazepines. Our objective is to create a resource that can enable the development of systems that can automatically detect clinical and social impacts of substance use from text-based social media data. The successful development of such systems may enable us to better understand how nonmedical use of substances affects individual health and societal dynamics, aiding the development of effective public health strategies. In addition to creating the annotated data set, we applied several machine learning models to establish baseline performances. Specifically, we experimented with transformer models like BERT, and RoBERTa, one few-shot learning model DANN by leveraging the full training dataset, and GPT-3.5 by using one-shot learning, for automatic NER of clinical and social impacts. The dataset has been made available through the 2024 SMM4H shared tasks.
We present a symbolic learning framework inspired by cognitive-like memory functionalities (i.e., storing, retrieving, consolidating and forgetting) to generate task representations to support high-level task planning and knowledge bootstrapping. We address a scenario involving a non-expert human, who performs a single task demonstration, and a robot, which online learns structured knowledge to re-execute the task based on experiences, i.e., observations. We consider a one-shot learning process based on non-annotated data to store an intelligible representation of the task, which can be refined through interaction, e.g., via verbal or visual communication. Our general-purpose framework relies on fuzzy Description Logic, which has been used to extend the previously developed Scene Identification and Tagging algorithm. In this paper, we exploit such an algorithm to implement cognitive-like memory functionalities employing scores that rank memorised observations over time based on simple heuristics. Our main contribution is the formalisation of a framework that can be used to systematically investigate different heuristics for bootstrapping hierarchical knowledge representations based on robot observations. Through an illustrative assembly task scenario, the paper presents the performance of our framework to discuss its benefits and limitations.
Dataset distillation aims at synthesizing a dataset by a small number of artificially generated data items, which, when used as training data, reproduce or approximate a machine learning (ML) model as if it were trained on the entire original dataset. Consequently, data distillation methods are usually tied to a specific ML algorithm. While recent literature deals mainly with distillation of large collections of images in the context of neural network models, tabular data distillation is much less represented and mainly focused on a theoretical perspective. The current paper explores the potential of a simple distillation technique previously proposed in the context of Less-than-one shot learning. The main goal is to push further the performance of prototype-based soft-labels distillation in terms of classification accuracy, by integrating optimization steps in the distillation process. The analysis is performed on real-world data sets with various degrees of imbalance. Experimental studies trace the capability of the method to distill the data, but also the opportunity to act as an augmentation method, i.e. to generate new data that is able to increase model accuracy when used in conjunction with - as opposed to instead of - the original data.
A hallmark of human intelligence is the ability to infer abstract rules from limited experience and apply these rules to unfamiliar situations. This capacity is widely studied in the visual domain using the Raven's Progressive Matrices. Recent advances in deep learning have led to multiple artificial neural network models matching or even surpassing human performance. However, while humans can identify and express the rule underlying these tasks with little to no exposure, contemporary neural networks often rely on massive pattern-based training and cannot express or extrapolate the rule inferred from the task. Furthermore, most Raven's Progressive Matrices or Raven-like tasks used for neural network training used symbolic representations, whereas humans can flexibly switch between symbolic and continuous perceptual representations. In this work, we present an algorithmic approach to rule detection and application using feature detection, affine transformation estimation and search. We applied our model to a simplified Raven's Progressive Matrices task, previously designed for behavioral testing and neuroimaging in humans. The model exhibited one-shot learning and achieved near human-level performance in the symbolic reasoning condition of the simplified task. Furthermore, the model can express the relationships discovered and generate multi-step predictions in accordance with the underlying rule. Finally, the model can reason using continuous patterns. We discuss our results and their relevance to studying abstract reasoning in humans, as well as their implications for improving intelligent machines.
The advent of Large Language Models (LLMs) has advanced the benchmark in various Natural Language Processing (NLP) tasks. However, large amounts of labelled training data are required to train LLMs. Furthermore, data annotation and training are computationally expensive and time-consuming. Zero and few-shot learning have recently emerged as viable options for labelling data using large pre-trained models. Hate speech detection in mix-code low-resource languages is an active problem area where the use of LLMs has proven beneficial. In this study, we have compiled a dataset of 100 YouTube comments, and weakly labelled them for coarse and fine-grained misogyny classification in mix-code Hinglish. Weak annotation was applied due to the labor-intensive annotation process. Zero-shot learning, one-shot learning, and few-shot learning and prompting approaches have then been applied to assign labels to the comments and compare them to human-assigned labels. Out of all the approaches, zero-shot classification using the Bidirectional Auto-Regressive Transformers (BART) large model and few-shot prompting using Generative Pre-trained Transformer- 3 (ChatGPT-3) achieve the best results
Semantic segmentation is a key prerequisite to robust image understanding for applications in \acrlong{ai} and Robotics. \acrlong{fss}, in particular, concerns the extension and optimization of traditional segmentation methods in challenging conditions where limited training examples are available. A predominant approach in \acrlong{fss} is to rely on a single backbone for visual feature extraction. Choosing which backbone to leverage is a deciding factor contributing to the overall performance. In this work, we interrogate on whether fusing features from different backbones can improve the ability of \acrlong{fss} models to capture richer visual features. To tackle this question, we propose and compare two ensembling techniques-Independent Voting and Feature Fusion. Among the available \acrlong{fss} methods, we implement the proposed ensembling techniques on PANet. The module dedicated to predicting segmentation masks from the backbone embeddings in PANet avoids trainable parameters, creating a controlled `in vitro' setting for isolating the impact of different ensembling strategies. Leveraging the complementary strengths of different backbones, our approach outperforms the original single-backbone PANet across standard benchmarks even in challenging one-shot learning scenarios. Specifically, it achieved a performance improvement of +7.37\% on PASCAL-5\textsuperscript{i} and of +10.68\% on COCO-20\textsuperscript{i} in the top-performing scenario where three backbones are combined. These results, together with the qualitative inspection of the predicted subject masks, suggest that relying on multiple backbones in PANet leads to a more comprehensive feature representation, thus expediting the successful application of \acrlong{fss} methods in challenging, data-scarce environments.
The potential for augmenting the segmentation of brain tumors through the use of few-shot learning is vast. Although several deep learning networks (DNNs) demonstrate promising results in terms of segmentation, they require a substantial quantity of training data in order to produce suitable outcomes. Furthermore, a major issue faced by most of these models is their ability to perform well when faced with unseen classes. To address these challenges, we propose a one-shot learning model for segmenting brain tumors in magnetic resonance images (MRI) of the brain, based on a single prototype similarity score. Leveraging the recently developed techniques of few-shot learning, which involve the utilization of support and query sets of images for training and testing purposes, we strive to obtain a definitive tumor region by focusing on slices that contain foreground classes. This approach differs from other recent DNNs that utilize the entire set of images. The training process for this model is carried out iteratively, with each iteration involving the selection of random slices that contain foreground classes from randomly sampled data as the query set, along with a different random slice from the same sample as the support set. In order to distinguish the query images from the class prototypes, we employ a metric learning-based approach that relies on non-parametric thresholds. We employ the multimodal Brain Tumor Image Segmentation (BraTS) 2021 dataset, which comprises 60 training images and 350 testing images. The effectiveness of the model is assessed using the mean dice score and mean Intersection over Union (IoU) score.
In an era marked by a rapid increase in scientific publications, researchers grapple with the challenge of keeping pace with field-specific advances. We present the `AHAM' methodology and a metric that guides the domain-specific \textbf{adapt}ation of the BERTopic topic modeling framework to improve scientific text analysis. By utilizing the LLaMa2 generative language model, we generate topic definitions via one-shot learning by crafting prompts with the \textbf{help} of domain experts to guide the LLM for literature mining by \textbf{asking} it to model the topic names. For inter-topic similarity evaluation, we leverage metrics from language generation and translation processes to assess lexical and semantic similarity of the generated topics. Our system aims to reduce both the ratio of outlier topics to the total number of topics and the similarity between topic definitions. The methodology has been assessed on a newly gathered corpus of scientific papers on literature-based discovery. Through rigorous evaluation by domain experts, AHAM has been validated as effective in uncovering intriguing and novel insights within broad research areas. We explore the impact of domain adaptation of sentence-transformers for the task of topic \textbf{model}ing using two datasets, each specialized to specific scientific domains within arXiv and medarxiv. We evaluate the impact of data size, the niche of adaptation, and the importance of domain adaptation. Our results suggest a strong interaction between domain adaptation and topic modeling precision in terms of outliers and topic definitions.