German BERT Model for Legal Named Entity Recognition

The use of BERT, one of the most popular language models, has led to improvements in many Natural Language Processing (NLP) tasks. One such task is Named Entity Recognition (NER) i.e. automatic identification of named entities such as location, person, organization, etc. from a given text. It is also an important base step for many NLP tasks such as information extraction and argumentation mining. Even though there is much research done on NER using BERT and other popular language models, the same is not explored in detail when it comes to Legal NLP or Legal Tech. Legal NLP applies various NLP techniques such as sentence similarity or NER specifically on legal data. There are only a handful of models for NER tasks using BERT language models, however, none of these are aimed at legal documents in German. In this paper, we fine-tune a popular BERT language model trained on German data (German BERT) on a Legal Entity Recognition (LER) dataset. To make sure our model is not overfitting, we performed a stratified 10-fold cross-validation. The results we achieve by fine-tuning German BERT on the LER dataset outperform the BiLSTM-CRF+ model used by the authors of the same LER dataset. Finally, we make the model openly available via HuggingFace.

Data-Association-Free Landmark-based SLAM

We study landmark-based SLAM with unknown data association: our robot navigates in a completely unknown environment and has to simultaneously reason over its own trajectory, the positions of an unknown number of landmarks in the environment, and potential data associations between measurements and landmarks. This setup is interesting since: (i) it arises when recovering from data association failures or from SLAM with information-poor sensors, (ii) it sheds light on fundamental limits (and hardness) of landmark-based SLAM problems irrespective of the front-end data association method, and (iii) it generalizes existing approaches where data association is assumed to be known or partially known. We approach the problem by splitting it into an inner problem of estimating the trajectory, landmark positions and data associations and an outer problem of estimating the number of landmarks. Our approach creates useful and novel connections with existing techniques from discrete-continuous optimization (e.g., k-means clustering), which has the potential to trigger novel research. We demonstrate the proposed approaches in extensive simulations and on real datasets and show that the proposed techniques outperform typical data association baselines and are even competitive against an "oracle" baseline which has access to the number of landmarks and an initial guess for each landmark.

Introspective Cross-Attention Probing for Lightweight Transfer of Pre-trained Models

We propose InCA, a lightweight method for transfer learning that cross-attends to any activation layer of a pre-trained model. During training, InCA uses a single forward pass to extract multiple activations, which are passed to external cross-attention adapters, trained anew and combined or selected for downstream tasks. We show that, even when selecting a single top-scoring adapter, InCA achieves performance comparable to full fine-tuning, at a cost comparable to fine-tuning just the last layer. For example, with a cross-attention probe 1.3% the size of a pre-trained ViT-L/16 model, we achieve performance within 0.2% of the full fine-tuning paragon at 51% training cost of the baseline, on average across 11 downstream classification tasks. Unlike other forms of efficient adaptation, InCA does not require backpropagating through the pre-trained model, thus leaving its execution unaltered at both training and inference. The versatility of InCA is best illustrated in fine-grained tasks, which may require accessing information absent in the last layer but accessible in intermediate layer activations. Since the backbone is fixed, InCA allows parallel ensembling as well as parallel execution of multiple tasks. InCA achieves state-of-the-art performance in the ImageNet-to-Sketch multi-task benchmark.

A Privacy Preserving System for Movie Recommendations using Federated Learning

Recommender systems have become ubiquitous in the past years. They solve the tyranny of choice problem faced by many users, and are employed by many online businesses to drive engagement and sales. Besides other criticisms, like creating filter bubbles within social networks, recommender systems are often reproved for collecting considerable amounts of personal data. However, to personalize recommendations, personal information is fundamentally required. A recent distributed learning scheme called federated learning has made it possible to learn from personal user data without its central collection. Accordingly, we present a complete recommender system for movie recommendations, which provides privacy and thus trustworthiness on two levels: First, it is trained using federated learning and thus is, by its very nature, privacy-preserving, while still enabling individual users to benefit from global insights. And second, a novel federated learning scheme, FedQ, is employed, which not only addresses the problem of non-i.i.d. and small local datasets, but also prevents input data reconstruction attacks by aggregating client models early. To reduce the communication overhead, compression is applied, which significantly reduces the exchanged neural network updates to a fraction of their original data. We conjecture that it may also improve data privacy through its lossy quantization stage.

Exploring Long & Short Range Temporal Information for Learned Video Compression

Learned video compression methods have gained a variety of interest in the video coding community since they have matched or even exceeded the rate-distortion (RD) performance of traditional video codecs. However, many current learning-based methods are dedicated to utilizing short-range temporal information, thus limiting their performance. In this paper, we focus on exploiting the unique characteristics of video content and further exploring temporal information to enhance compression performance. Specifically, for long-range temporal information exploitation, we propose temporal prior that can update continuously within the group of pictures (GOP) during inference. In that case temporal prior contains valuable temporal information of all decoded images within the current GOP. As for short-range temporal information, we propose a progressive guided motion compensation to achieve robust and effective compensation. In detail, we design a hierarchical structure to achieve multi-scale compensation. More importantly, we use optical flow guidance to generate pixel offsets between feature maps at each scale, and the compensation results at each scale will be used to guide the following scale's compensation. Sufficient experimental results demonstrate that our method can obtain better RD performance than state-of-the-art video compression approaches. The code is publicly available on: https://github.com/Huairui/LSTVC.

Diffusion-Based Hierarchical Multi-Label Object Detection to Analyze Panoramic Dental X-rays

Due to the necessity for precise treatment planning, the use of panoramic X-rays to identify different dental diseases has tremendously increased. Although numerous ML models have been developed for the interpretation of panoramic X-rays, there has not been an end-to-end model developed that can identify problematic teeth with dental enumeration and associated diagnoses at the same time. To develop such a model, we structure the three distinct types of annotated data hierarchically following the FDI system, the first labeled with only quadrant, the second labeled with quadrant-enumeration, and the third fully labeled with quadrant-enumeration-diagnosis. To learn from all three hierarchies jointly, we introduce a novel diffusion-based hierarchical multi-label object detection framework by adapting a diffusion-based method that formulates object detection as a denoising diffusion process from noisy boxes to object boxes. Specifically, to take advantage of the hierarchically annotated data, our method utilizes a novel noisy box manipulation technique by adapting the denoising process in the diffusion network with the inference from the previously trained model in hierarchical order. We also utilize a multi-label object detection method to learn efficiently from partial annotations and to give all the needed information about each abnormal tooth for treatment planning. Experimental results show that our method significantly outperforms state-of-the-art object detection methods, including RetinaNet, Faster R-CNN, DETR, and DiffusionDet for the analysis of panoramic X-rays, demonstrating the great potential of our method for hierarchically and partially annotated datasets. The code and the data are available at: https://github.com/ibrahimethemhamamci/HierarchicalDet.

A Novel Method Combines Moving Fronts, Data Decomposition and Deep Learning to Forecast Intricate Time Series

A univariate time series with high variability can pose a challenge even to Deep Neural Network (DNN). To overcome this, a univariate time series is decomposed into simpler constituent series, whose sum equals the original series. As demonstrated in this article, the conventional one-time decomposition technique suffers from a leak of information from the future, referred to as a data leak. In this work, a novel Moving Front (MF) method is proposed to prevent data leakage, so that the decomposed series can be treated like other time series. Indian Summer Monsoon Rainfall (ISMR) is a very complex time series, which poses a challenge to DNN and is therefore selected as an example. From the many signal processing tools available, Empirical Wavelet Transform (EWT) was chosen for decomposing the ISMR into simpler constituent series, as it was found to be more effective than the other popular algorithm, Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN). The proposed MF method was used to generate the constituent leakage-free time series. Predictions and forecasts were made by state-of-the-art Long and Short-Term Memory (LSTM) network architecture, especially suitable for making predictions of sequential patterns. The constituent MF series has been divided into training, testing, and forecasting. It has been found that the model (EWT-MF-LSTM) developed here made exceptionally good train and test predictions, as well as Walk-Forward Validation (WFV), forecasts with Performance Parameter ($PP$) values of 0.99, 0.86, and 0.95, respectively, where $PP$ = 1.0 signifies perfect reproduction of the data.

Finetuning Pretrained Vision-Language Models with Correlation Information Bottleneck for Robust Visual Question Answering

Benefiting from large-scale Pretrained Vision-Language Models (VL-PMs), the performance of Visual Question Answering (VQA) has started to approach human oracle performance. However, finetuning large-scale VL-PMs with limited data for VQA usually faces overfitting and poor generalization issues, leading to a lack of robustness. In this paper, we aim to improve the robustness of VQA systems (ie, the ability of the systems to defend against input variations and human-adversarial attacks) from the perspective of Information Bottleneck when finetuning VL-PMs for VQA. Generally, internal representations obtained by VL-PMs inevitably contain irrelevant and redundant information for the downstream VQA task, resulting in statistically spurious correlations and insensitivity to input variations. To encourage representations to converge to a minimal sufficient statistic in vision-language learning, we propose the Correlation Information Bottleneck (CIB) principle, which seeks a tradeoff between representation compression and redundancy by minimizing the mutual information (MI) between the inputs and internal representations while maximizing the MI between the outputs and the representations. Meanwhile, CIB measures the internal correlations among visual and linguistic inputs and representations by a symmetrized joint MI estimation. Extensive experiments on five VQA benchmarks of input robustness and two VQA benchmarks of human-adversarial robustness demonstrate the effectiveness and superiority of the proposed CIB in improving the robustness of VQA systems.

Time Series Forecasting via Semi-Asymmetric Convolutional Architecture with Global Atrous Sliding Window

The proposed method in this paper is designed to address the problem of time series forecasting. Although some exquisitely designed models achieve excellent prediction performances, how to extract more useful information and make accurate predictions is still an open issue. Most of modern models only focus on a short range of information, which are fatal for problems such as time series forecasting which needs to capture long-term information characteristics. As a result, the main concern of this work is to further mine relationship between local and global information contained in time series to produce more precise predictions. In this paper, to satisfactorily realize the purpose, we make three main contributions that are experimentally verified to have performance advantages. Firstly, original time series is transformed into difference sequence which serves as input to the proposed model. And secondly, we introduce the global atrous sliding window into the forecasting model which references the concept of fuzzy time series to associate relevant global information with temporal data within a time period and utilizes central-bidirectional atrous algorithm to capture underlying-related features to ensure validity and consistency of captured data. Thirdly, a variation of widely-used asymmetric convolution which is called semi-asymmetric convolution is devised to more flexibly extract relationships in adjacent elements and corresponding associated global features with adjustable ranges of convolution on vertical and horizontal directions. The proposed model in this paper achieves state-of-the-art on most of time series datasets provided compared with competitive modern models.

Patch Network for medical image Segmentation

Accurate and fast segmentation of medical images is clinically essential, yet current research methods include convolutional neural networks with fast inference speed but difficulty in learning image contextual features, and transformer with good performance but high hardware requirements. In this paper, we present a Patch Network (PNet) that incorporates the Swin Transformer notion into a convolutional neural network, allowing it to gather richer contextual information while achieving the balance of speed and accuracy. We test our PNet on Polyp(CVC-ClinicDB and ETIS- LaribPolypDB), Skin(ISIC-2018 Skin lesion segmentation challenge dataset) segmentation datasets. Our PNet achieves SOTA performance in both speed and accuracy.