Watermark text spotting in document images can offer access to an often unexplored source of information, providing crucial evidence about a record's scope, audience and sometimes even authenticity. Stemming from the problem of text spotting, detecting and understanding watermarks in documents inherits the same hardships - in the wild, writing can come in various fonts, sizes and forms, making generic recognition a very difficult problem. To address the lack of resources in this field and propel further research, we propose a novel benchmark (K-Watermark) containing 65,447 data samples generated using Wrender, a watermark text patterns rendering procedure. A validity study using humans raters yields an authenticity score of 0.51 against pre-generated watermarked documents. To prove the usefulness of the dataset and rendering technique, we developed an end-to-end solution (Wextract) for detecting the bounding box instances of watermark text, while predicting the depicted text. To deal with this specific task, we introduce a variance minimization loss and a hierarchical self-attention mechanism. To the best of our knowledge, we are the first to propose an evaluation benchmark and a complete solution for retrieving watermarks from documents surpassing baselines by 5 AP points in detection and 4 points in character accuracy.
相關內容
Recent approaches to improving the extraction of text embeddings from autoregressive large language models (LLMs) have largely focused on improvements to data, backbone pretrained language models, or improving task-differentiation via instructions. In this work, we address an architectural limitation of autoregressive models: token embeddings cannot contain information from tokens that appear later in the input. To address this limitation, we propose a simple approach, "echo embeddings," in which we repeat the input twice in context and extract embeddings from the second occurrence. We show that echo embeddings of early tokens can encode information about later tokens, allowing us to maximally leverage high-quality LLMs for embeddings. On the MTEB leaderboard, echo embeddings improve over classical embeddings by over 9% zero-shot and by around 0.7% when fine-tuned. Echo embeddings with a Mistral-7B model achieve state-of-the-art compared to prior open source models that do not leverage synthetic fine-tuning data.
Controllable video editing has demonstrated remarkable potential across diverse applications, particularly in scenarios where capturing or re-capturing real-world videos is either impractical or costly. This paper introduces a novel and efficient system named Place-Anything, which facilitates the insertion of any object into any video solely based on a picture or text description of the target object or element. The system comprises three modules: 3D generation, video reconstruction, and 3D target insertion. This integrated approach offers an efficient and effective solution for producing and editing high-quality videos by seamlessly inserting realistic objects. Through a user study, we demonstrate that our system can effortlessly place any object into any video using just a photograph of the object. Our demo video can be found at //youtu.be/afXqgLLRnTE. Please also visit our project page //place-anything.github.io to get access.
Rapid progress in scalable, commoditized tools for data collection and data processing has made it possible for firms and policymakers to employ ever more complex metrics as guides for decision-making. These developments have highlighted a prevailing challenge -- deciding *which* metrics to compute. In particular, a firm's ability to compute a wider range of existing metrics does not address the problem of *unknown unknowns*, which reflects informational limitations on the part of the firm. To guide the choice of metrics in the face of this informational problem, we turn to the evaluated agents themselves, who may have more information than a principal about how to measure outcomes effectively. We model this interaction as a simple agency game, where we ask: *When does an agent have an incentive to reveal the observability of a cost-correlated variable to the principal?* There are two effects: better information reduces the agent's information rents but also makes some projects go forward that otherwise would fail. We show that the agent prefers to reveal information that exposes a strong enough differentiation between high and low costs. Expanding the agent's action space to include the ability to *garble* their information, we show that the agent often prefers to garble over full revelation. Still, giving the agent the ability to garble can lead to higher total welfare. Our model has analogies with price discrimination, and we leverage some of these synergies to analyze total welfare.
Matrix factorization (MF) is a simple collaborative filtering technique that achieves superior recommendation accuracy by decomposing the user-item interaction matrix into user and item latent matrices. Because the model typically learns each interaction independently, it may overlook the underlying shared dependencies between users and items, resulting in less stable and interpretable recommendations. Based on these insights, we propose "Hierarchical Matrix Factorization" (HMF), which incorporates clustering concepts to capture the hierarchy, where leaf nodes and other nodes correspond to users/items and clusters, respectively. Central to our approach, called hierarchical embeddings, is the additional decomposition of the latent matrices (embeddings) into probabilistic connection matrices, which link the hierarchy, and a root cluster latent matrix. The embeddings are differentiable, allowing simultaneous learning of interactions and clustering using a single gradient descent method. Furthermore, the obtained cluster-specific interactions naturally summarize user-item interactions and provide interpretability. Experimental results on ratings and ranking predictions show that HMF outperforms existing MF methods, in particular achieving a 1.37 point improvement in RMSE for sparse interactions. Additionally, it was confirmed that the clustering integration of HMF has the potential for faster learning convergence and mitigation of overfitting compared to MF, and also provides interpretability through a cluster-centered case study.
Context and motivation: Requirements engineering of complex IT systems needs to manage the many, and often vague and conflicting, organisational rules that exist in the context of a modern enterprise. At the same time, IT systems affect the organisation, essentially setting new rules on how the organisation should work. Question/problem: Gathering requirements for an IT system involves understanding the complex rules that govern an organisation. The research question is: How can the holistic properties of organisational rules be conceptualised? Principal ideas/results: This paper introduces the concept of organisational rule systems that may be used to describe complex organisational rules. The concept and its components are presented as a conceptual framework, which in turn is condensed into a conceptual framework diagram. The framework is grounded in a critical literature review. Contribution: The conceptual framework will, as a first step of a wider research agenda, help requirements engineers understand the influence of organisational rules.
Sequence labeling models often benefit from incorporating external knowledge. However, this practice introduces data heterogeneity and complicates the model with additional modules, leading to increased expenses for training a high-performing model. To address this challenge, we propose a two-stage curriculum learning (TCL) framework specifically designed for sequence labeling tasks. The TCL framework enhances training by gradually introducing data instances from easy to hard, aiming to improve both performance and training speed. Furthermore, we explore different metrics for assessing the difficulty levels of sequence labeling tasks. Through extensive experimentation on six Chinese word segmentation (CWS) and Part-of-speech tagging (POS) datasets, we demonstrate the effectiveness of our model in enhancing the performance of sequence labeling models. Additionally, our analysis indicates that TCL accelerates training and alleviates the slow training problem associated with complex models.
Diffusion models generate highly realistic images by learning a multi-step denoising process, naturally embodying the principles of multi-task learning (MTL). Despite the inherent connection between diffusion models and MTL, there remains an unexplored area in designing neural architectures that explicitly incorporate MTL into the framework of diffusion models. In this paper, we present Denoising Task Routing (DTR), a simple add-on strategy for existing diffusion model architectures to establish distinct information pathways for individual tasks within a single architecture by selectively activating subsets of channels in the model. What makes DTR particularly compelling is its seamless integration of prior knowledge of denoising tasks into the framework: (1) Task Affinity: DTR activates similar channels for tasks at adjacent timesteps and shifts activated channels as sliding windows through timesteps, capitalizing on the inherent strong affinity between tasks at adjacent timesteps. (2) Task Weights: During the early stages (higher timesteps) of the denoising process, DTR assigns a greater number of task-specific channels, leveraging the insight that diffusion models prioritize reconstructing global structure and perceptually rich contents in earlier stages, and focus on simple noise removal in later stages. Our experiments reveal that DTR not only consistently boosts diffusion models' performance across different evaluation protocols without adding extra parameters but also accelerates training convergence. Finally, we show the complementarity between our architectural approach and existing MTL optimization techniques, providing a more complete view of MTL in the context of diffusion training. Significantly, by leveraging this complementarity, we attain matched performance of DiT-XL using the smaller DiT-L with a reduction in training iterations from 7M to 2M.
Recent advances in maximizing mutual information (MI) between the source and target have demonstrated its effectiveness in text generation. However, previous works paid little attention to modeling the backward network of MI (i.e., dependency from the target to the source), which is crucial to the tightness of the variational information maximization lower bound. In this paper, we propose Adversarial Mutual Information (AMI): a text generation framework which is formed as a novel saddle point (min-max) optimization aiming to identify joint interactions between the source and target. Within this framework, the forward and backward networks are able to iteratively promote or demote each other's generated instances by comparing the real and synthetic data distributions. We also develop a latent noise sampling strategy that leverages random variations at the high-level semantic space to enhance the long term dependency in the generation process. Extensive experiments based on different text generation tasks demonstrate that the proposed AMI framework can significantly outperform several strong baselines, and we also show that AMI has potential to lead to a tighter lower bound of maximum mutual information for the variational information maximization problem.
The notion of "in-domain data" in NLP is often over-simplistic and vague, as textual data varies in many nuanced linguistic aspects such as topic, style or level of formality. In addition, domain labels are many times unavailable, making it challenging to build domain-specific systems. We show that massive pre-trained language models implicitly learn sentence representations that cluster by domains without supervision -- suggesting a simple data-driven definition of domains in textual data. We harness this property and propose domain data selection methods based on such models, which require only a small set of in-domain monolingual data. We evaluate our data selection methods for neural machine translation across five diverse domains, where they outperform an established approach as measured by both BLEU and by precision and recall of sentence selection with respect to an oracle.
Knowledge graph (KG) embedding encodes the entities and relations from a KG into low-dimensional vector spaces to support various applications such as KG completion, question answering, and recommender systems. In real world, knowledge graphs (KGs) are dynamic and evolve over time with addition or deletion of triples. However, most existing models focus on embedding static KGs while neglecting dynamics. To adapt to the changes in a KG, these models need to be re-trained on the whole KG with a high time cost. In this paper, to tackle the aforementioned problem, we propose a new context-aware Dynamic Knowledge Graph Embedding (DKGE) method which supports the embedding learning in an online fashion. DKGE introduces two different representations (i.e., knowledge embedding and contextual element embedding) for each entity and each relation, in the joint modeling of entities and relations as well as their contexts, by employing two attentive graph convolutional networks, a gate strategy, and translation operations. This effectively helps limit the impacts of a KG update in certain regions, not in the entire graph, so that DKGE can rapidly acquire the updated KG embedding by a proposed online learning algorithm. Furthermore, DKGE can also learn KG embedding from scratch. Experiments on the tasks of link prediction and question answering in a dynamic environment demonstrate the effectiveness and efficiency of DKGE.
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