Recent advancements in Large Language Models (LLMs) have significantly improved their performance across various Natural Language Processing (NLP) tasks. However, LLMs still struggle with generating non-factual responses due to limitations in their parametric memory. Retrieval-Augmented Generation (RAG) systems address this issue by incorporating external knowledge with a retrieval module. Despite their successes, however, current RAG systems face challenges with retrieval failures and the limited ability of LLMs to filter out irrelevant information. Therefore, in this work, we propose DSLR (Document Refinement with Sentence-Level Re-ranking and Reconstruction), an unsupervised framework that decomposes retrieved documents into sentences, filters out irrelevant sentences, and reconstructs them again into coherent passages. We experimentally validate DSLR on multiple open-domain QA datasets and the results demonstrate that DSLR significantly enhances the RAG performance over conventional fixed-size passage. Furthermore, our DSLR enhances performance in specific, yet realistic scenarios without the need for additional training, providing an effective and efficient solution for refining retrieved documents in RAG systems.

We investigate the replay buffer in rehearsal-based approaches for graph continual learning (GCL) methods. Existing rehearsal-based GCL methods select the most representative nodes for each class and store them in a replay buffer for later use in training subsequent tasks. However, we discovered that considering only the class representativeness of each replayed node makes the replayed nodes to be concentrated around the center of each class, incurring a potential risk of overfitting to nodes residing in those regions, which aggravates catastrophic forgetting. Moreover, as the rehearsal-based approach heavily relies on a few replayed nodes to retain knowledge obtained from previous tasks, involving the replayed nodes that have irrelevant neighbors in the model training may have a significant detrimental impact on model performance. In this paper, we propose a GCL model named DSLR, specifically, we devise a coverage-based diversity (CD) approach to consider both the class representativeness and the diversity within each class of the replayed nodes. Moreover, we adopt graph structure learning (GSL) to ensure that the replayed nodes are connected to truly informative neighbors. Extensive experimental results demonstrate the effectiveness and efficiency of DSLR. Our source code is available at https://github.com/seungyoon-Choi/DSLR_official.

We investigate a new paradigm that uses differentiable SLAM architectures in a self-supervised manner to train end-to-end deep learning models in various LiDAR based applications. To the best of our knowledge there does not exist any work that leverages SLAM as a training signal for deep learning based models. We explore new ways to improve the efficiency, robustness, and adaptability of LiDAR systems with deep learning techniques. We focus on the potential benefits of differentiable SLAM architectures for improving performance of deep learning tasks such as classification, regression as well as SLAM. Our experimental results demonstrate a non-trivial increase in the performance of two deep learning applications - Ground Level Estimation and Dynamic to Static LiDAR Translation, when used with differentiable SLAM architectures. Overall, our findings provide important insights that enhance the performance of LiDAR based navigation systems. We demonstrate that this new paradigm of using SLAM Loss signal while training LiDAR based models can be easily adopted by the community.

The quality assessment (QA) of restored low light images is an important tool for benchmarking and improving low light restoration (LLR) algorithms. While several LLR algorithms exist, the subjective perception of the restored images has been much less studied. Challenges in capturing aligned low light and well-lit image pairs and collecting a large number of human opinion scores of quality for training, warrant the design of unsupervised (or opinion unaware) no-reference (NR) QA methods. This work studies the subjective perception of low light restored images and their unsupervised NR QA. Our contributions are two-fold. We first create a dataset of restored low light images using various LLR methods, conduct a subjective QA study and benchmark the performance of existing QA methods. We then present a self-supervised contrastive learning technique to extract distortion aware features from the restored low light images. We show that these features can be effectively used to build an opinion unaware image quality analyzer. Detailed experiments reveal that our unsupervised NR QA model achieves state-of-the-art performance among all such quality measures for low light restored images.

Systems whose entities interact with each other are common. In many interacting systems, it is difficult to observe the relations between entities which is the key information for analyzing the system. In recent years, there has been increasing interest in discovering the relationships between entities using graph neural networks. However, existing approaches are difficult to apply if the number of relations is unknown or if the relations are complex. We propose the DiScovering Latent Relation (DSLR) model, which is flexibly applicable even if the number of relations is unknown or many types of relations exist. The flexibility of our DSLR model comes from the design concept of our encoder that represents the relation between entities in a latent space rather than a discrete variable and a decoder that can handle many types of relations. We performed the experiments on synthetic and real-world graph data with various relationships between entities, and compared the qualitative and quantitative results with other approaches. The experiments show that the proposed method is suitable for analyzing dynamic graphs with an unknown number of complex relations.

A couple of years ago I wrote about Google's AI camera, Google Clips ($250). This is a device that you plonk on the kitchen table (or hang around your neck), and which automatically takes photos whenever your favourite landscape, child or pet steps into frame. I compared it with a Nikon DSLR ($3,300) and proclaimed that AI devices would consume the lower end of the market while creative photographers would cling on to their interchangeable lenses. On reflection I'm not surprised by the news this week that Google Clips has been withdrawn. As we all know, AI relies heavily on machine learning, which requires huge volumes of data and experiments to accurately predict everything from eye disease to your next favourite artist on Spotify. Google tried to teach its camera about composition, subject focus and other skills using photo libraries but even then the results were disappointing.

Digital photographers coined the term "pixel peepers" years ago to denote -- mostly with scorn -- people who focused on flaws in the individual dots that create photos rather than the entirety of the images. Zooming in to 100%, it was said, is nothing but a recipe for perpetual disappointment; instead, judge each camera by the overall quality of the photo it takes, and don't get too mired in the details. Until now, Apple's approach to digital photography has been defined by its commitment to improving the quality of the big picture without further compromising pixel-level quality. I say "further" because there's no getting around the fact that tiny phone camera sensors are physically incapable of matching the pixel-level results of full-frame DSLR camera sensors in a fair fight. Bigger sensors can capture more light and almost invariably more actual pixels than the iPhone's 12-megapixel cameras.

It's given us all a chance to save our memories, and to relive them when we see them again in our photos. That technology has come quite a long way over the past several years. With all kinds of new features like 4K, HDR, and colour enhancement, the photos one can capture are awe-inspiring. But it does come at a price. Not everyone can afford the best-of-the-best camera.

Artificial Intelligence (AI) is hardly in its infancy stage. It is not that accessible and thus the AI based solutions that we are using or are being deployed are far inferior to what we expect in the next two to three decades. We have experienced quite a lot of users express their frustration on Chatbots. They are built based on decision-tree logic, algorithms and thought concepts, where the response given by the bot depends on specific keywords that are identified in the user's input. Now if the user's input contains'DSLR' then a Chatbot would often send a message with a never-ending product list.

Smartphone cameras are pretty incredible things to have in your pocket, and the Pixel 2 does a very good job of making every image look fantastic. But you can't do better than a big, full-frame DSLR – the trouble is, they're not very pocket-friendly. So, if you're fed up of your phone taking washed out, shallow photos, this AI is designed to take your old smartphone pictures and give them DSLR-like quality – even if your smartphone isn't all that snazzy. Known as WESPE (Weakly Supervised Photo Enhancer), the team of data scientists behind the project aim to bring DSLR-like qualities to smartphone cameras. The idea is that, by training a deep learning system using photos of the same scene taken with a phone camera and on a DSLR, it'll learn the difference and automatically make those adjustments on images it's never seen before.