Scene text removal (STR) aims at replacing text strokes in natural scenes with visually coherent backgrounds. Recent STR approaches rely on iterative refinements or explicit text masks, resulting in higher complexity and sensitivity to the accuracy of text localization. Moreover, most existing STR methods utilize convolutional neural networks (CNNs) for feature representation while the potential of vision Transformers (ViTs) remains largely unexplored. In this paper, we propose a simple-yet-effective ViT-based text eraser, dubbed ViTEraser. Following a concise encoder-decoder framework, different types of ViTs can be easily integrated into ViTEraser to enhance the long-range dependencies and global reasoning. Specifically, the encoder hierarchically maps the input image into the hidden space through ViT blocks and patch embedding layers, while the decoder gradually upsamples the hidden features to the text-erased image with ViT blocks and patch splitting layers. As ViTEraser implicitly integrates text localization and inpainting, we propose a novel end-to-end pretraining method, termed SegMIM, which focuses the encoder and decoder on the text box segmentation and masked image modeling tasks, respectively. To verify the effectiveness of the proposed methods, we comprehensively explore the architecture, pretraining, and scalability of the ViT-based encoder-decoder for STR, which provides deep insights into the application of ViT to STR. Experimental results demonstrate that ViTEraser with SegMIM achieves state-of-the-art performance on STR by a substantial margin. Furthermore, the extended experiment on tampered scene text detection demonstrates the generality of ViTEraser to other tasks. We believe this paper can inspire more research on ViT-based STR approaches. Code will be available at //github.com/shannanyinxiang/ViTEraser.
相關內容
Foundation language models obtain the instruction-following ability through supervised fine-tuning (SFT). Diversity and complexity are considered critical factors of a successful SFT dataset, while their definitions remain obscure and lack quantitative analyses. In this work, we propose InsTag, an open-set fine-grained tagger, to tag samples within SFT datasets based on semantics and intentions and define instruction diversity and complexity regarding tags. We obtain 6.6K tags to describe comprehensive user queries. Then we analyze popular open-sourced SFT datasets and find that the model ability grows with more diverse and complex data. Based on this observation, we propose a data selector based on InsTag to select 6K diverse and complex samples from open-source datasets and fine-tune models on InsTag-selected data. The resulting models, TagLM, outperform open-source models based on considerably larger SFT data evaluated by MT-Bench, echoing the importance of query diversity and complexity. We open-source InsTag in //github.com/OFA-Sys/InsTag.
Label Distribution Learning (LDL) assigns soft labels, a.k.a. degrees, to a sample. In reality, it is always laborious to obtain complete degrees, giving birth to the Incomplete LDL (InLDL). However, InLDL often suffers from performance degeneration. To remedy it, existing methods need one or more explicit regularizations, leading to burdensome parameter tuning and extra computation. We argue that label distribution itself may provide useful prior, when used appropriately, the InLDL problem can be solved without any explicit regularization. In this paper, we offer a rational alternative to use such a prior. Our intuition is that large degrees are likely to get more concern, the small ones are easily overlooked, whereas the missing degrees are completely neglected in InLDL. To learn an accurate label distribution, it is crucial not to ignore the small observed degrees but to give them properly large weights, while gradually increasing the weights of the missing degrees. To this end, we first define a weighted empirical risk and derive upper bounds between the expected risk and the weighted empirical risk, which reveals in principle that weighting plays an implicit regularization role. Then, by using the prior of degrees, we design a weighted scheme and verify its effectiveness. To sum up, our model has four advantages, it is 1) model selection free, as no explicit regularization is imposed; 2) with closed form solution (sub-problem) and easy-to-implement (a few lines of codes); 3) with linear computational complexity in the number of samples, thus scalable to large datasets; 4) competitive with state-of-the-arts even without any explicit regularization.
Recently developed large language models have achieved remarkable success in generating fluent and coherent text. However, these models often tend to 'hallucinate' which critically hampers their reliability. In this work, we address this crucial problem and propose an approach that actively detects and mitigates hallucinations during the generation process. Specifically, we first identify the candidates of potential hallucination leveraging the model's logit output values, check their correctness through a validation procedure, mitigate the detected hallucinations, and then continue with the generation process. Through extensive experiments with GPT-3.5 (text-davinci-003) on the 'article generation task', we first demonstrate the individual efficacy of our detection and mitigation techniques. Specifically, the detection technique achieves a recall of ~88% and the mitigation technique successfully mitigates 57.6% of the correctly detected hallucinations. Importantly, our mitigation technique does not introduce new hallucinations even in the case of incorrectly detected hallucinations, i.e., false positives. Then, we show that the proposed active detection and mitigation approach successfully reduces the hallucinations of the GPT-3.5 model from 47.5% to 14.5% on average. We further demonstrate the effectiveness and wide applicability of our approach through additional studies including performance on different types of questions (multi-hop and false premise questions) and with another LLM from a different model family (Vicuna). In summary, our work contributes to improving the reliability and trustworthiness of large language models, a crucial step en route to enabling their widespread adoption in real-world applications.
We present CLASSLA-Stanza, a pipeline for automatic linguistic annotation of the South Slavic languages, which is based on the Stanza natural language processing pipeline. We describe the main improvements in CLASSLA-Stanza with respect to Stanza, and give a detailed description of the model training process for the latest 2.1 release of the pipeline. We also report performance scores produced by the pipeline for different languages and varieties. CLASSLA-Stanza exhibits consistently high performance across all the supported languages and outperforms or expands its parent pipeline Stanza at all the supported tasks. We also present the pipeline's new functionality enabling efficient processing of web data and the reasons that led to its implementation.
Mentions of new concepts appear regularly in texts and require automated approaches to harvest and place them into Knowledge Bases (KB), e.g., ontologies and taxonomies. Existing datasets suffer from three issues, (i) mostly assuming that a new concept is pre-discovered and cannot support out-of-KB mention discovery; (ii) only using the concept label as the input along with the KB and thus lacking the contexts of a concept label; and (iii) mostly focusing on concept placement w.r.t a taxonomy of atomic concepts, instead of complex concepts, i.e., with logical operators. To address these issues, we propose a new benchmark, adapting MedMentions dataset (PubMed abstracts) with SNOMED CT versions in 2014 and 2017 under the Diseases sub-category and the broader categories of Clinical finding, Procedure, and Pharmaceutical / biologic product. We provide usage on the evaluation with the dataset for out-of-KB mention discovery and concept placement, adapting recent Large Language Model based methods.
Discovering entity mentions that are out of a Knowledge Base (KB) from texts plays a critical role in KB maintenance, but has not yet been fully explored. The current methods are mostly limited to the simple threshold-based approach and feature-based classification, and the datasets for evaluation are relatively rare. We propose BLINKout, a new BERT-based Entity Linking (EL) method which can identify mentions that do not have corresponding KB entities by matching them to a special NIL entity. To better utilize BERT, we propose new techniques including NIL entity representation and classification, with synonym enhancement. We also apply KB Pruning and Versioning strategies to automatically construct out-of-KB datasets from common in-KB EL datasets. Results on five datasets of clinical notes, biomedical publications, and Wikipedia articles in various domains show the advantages of BLINKout over existing methods to identify out-of-KB mentions for the medical ontologies, UMLS, SNOMED CT, and the general KB, WikiData.
We present VERF, a collection of two methods (VERF-PnP and VERF-Light) for providing runtime assurance on the correctness of a camera pose estimate of a monocular camera without relying on direct depth measurements. We leverage the ability of NeRF (Neural Radiance Fields) to render novel RGB perspectives of a scene. We only require as input the camera image whose pose is being estimated, an estimate of the camera pose we want to monitor, and a NeRF model containing the scene pictured by the camera. We can then predict if the pose estimate is within a desired distance from the ground truth and justify our prediction with a level of confidence. VERF-Light does this by rendering a viewpoint with NeRF at the estimated pose and estimating its relative offset to the sensor image up to scale. Since scene scale is unknown, the approach renders another auxiliary image and reasons over the consistency of the optical flows across the three images. VERF-PnP takes a different approach by rendering a stereo pair of images with NeRF and utilizing the Perspective-n-Point (PnP) algorithm. We evaluate both methods on the LLFF dataset, on data from a Unitree A1 quadruped robot, and on data collected from Blue Origin's sub-orbital New Shepard rocket to demonstrate the effectiveness of the proposed pose monitoring method across a range of scene scales. We also show monitoring can be completed in under half a second on a 3090 GPU.
Dimensionality reduction (DR) techniques inherently distort the original structure of input high-dimensional data, producing imperfect low-dimensional embeddings. Diverse distortion measures have thus been proposed to evaluate the reliability of DR embeddings. However, implementing and executing distortion measures in practice has so far been time-consuming and tedious. To address this issue, we present ZADU, a Python library that provides distortion measures. ZADU is not only easy to install and execute but also enables comprehensive evaluation of DR embeddings through three key features. First, the library covers a wide range of distortion measures. Second, it automatically optimizes the execution of distortion measures, substantially reducing the running time required to execute multiple measures. Last, the library informs how individual points contribute to the overall distortions, facilitating the detailed analysis of DR embeddings. By simulating a real-world scenario of optimizing DR embeddings, we verify that our optimization scheme substantially reduces the time required to execute distortion measures. Finally, as an application of ZADU, we present another library called ZADUVis that allows users to easily create distortion visualizations that depict the extent to which each region of an embedding suffers from distortions.
We present a novel method, called NeTO, for capturing 3D geometry of solid transparent objects from 2D images via volume rendering. Reconstructing transparent objects is a very challenging task, which is ill-suited for general-purpose reconstruction techniques due to the specular light transport phenomena. Although existing refraction-tracing based methods, designed specially for this task, achieve impressive results, they still suffer from unstable optimization and loss of fine details, since the explicit surface representation they adopted is difficult to be optimized, and the self-occlusion problem is ignored for refraction-tracing. In this paper, we propose to leverage implicit Signed Distance Function (SDF) as surface representation, and optimize the SDF field via volume rendering with a self-occlusion aware refractive ray tracing. The implicit representation enables our method to be capable of reconstructing high-quality reconstruction even with a limited set of images, and the self-occlusion aware strategy makes it possible for our method to accurately reconstruct the self-occluded regions. Experiments show that our method achieves faithful reconstruction results and outperforms prior works by a large margin. Visit our project page at \url{//www.xxlong.site/NeTO/}
Collecting supporting evidence from large corpora of text (e.g., Wikipedia) is of great challenge for open-domain Question Answering (QA). Especially, for multi-hop open-domain QA, scattered evidence pieces are required to be gathered together to support the answer extraction. In this paper, we propose a new retrieval target, hop, to collect the hidden reasoning evidence from Wikipedia for complex question answering. Specifically, the hop in this paper is defined as the combination of a hyperlink and the corresponding outbound link document. The hyperlink is encoded as the mention embedding which models the structured knowledge of how the outbound link entity is mentioned in the textual context, and the corresponding outbound link document is encoded as the document embedding representing the unstructured knowledge within it. Accordingly, we build HopRetriever which retrieves hops over Wikipedia to answer complex questions. Experiments on the HotpotQA dataset demonstrate that HopRetriever outperforms previously published evidence retrieval methods by large margins. Moreover, our approach also yields quantifiable interpretations of the evidence collection process.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191