Scene Graph Generation (SGG) is a task that encodes visual relationships between objects in images as graph structures. SGG shows significant promise as a foundational component for downstream tasks, such as reasoning for embodied agents. To enable real-time applications, SGG must address the trade-off between performance and inference speed. However, current methods tend to focus on one of the following: (1) improving relation prediction accuracy, (2) enhancing object detection accuracy, or (3) reducing latency, without aiming to balance all three objectives simultaneously. To address this limitation, we propose a novel architecture, inference method, and relation prediction model. Our proposed solution, the REACT model, achieves the highest inference speed among existing SGG models, improving object detection accuracy without sacrificing relation prediction performance. Compared to state-of-the-art approaches, REACT is 2.7 times faster (with a latency of 23 ms) and improves object detection accuracy by 58.51%. Furthermore, our proposal significantly reduces model size, with an average of 5.5x fewer parameters. Code is available at https://github.com/Maelic/SGG-Benchmark
Document layout analysis (DLA) is crucial for understanding the physical layout and logical structure of documents, serving information retrieval, document summarization, knowledge extraction, etc. However, previous studies have typically used separate models to address individual sub-tasks within DLA, including table/figure detection, text region detection, logical role classification, and reading order prediction. In this work, we propose an end-to-end transformer-based approach for document layout analysis, called DLAFormer, which integrates all these sub-tasks into a single model. To achieve this, we treat various DLA sub-tasks (such as text region detection, logical role classification, and reading order prediction) as relation prediction problems and consolidate these relation prediction labels into a unified label space, allowing a unified relation prediction module to handle multiple tasks concurrently. Additionally, we introduce a novel set of type-wise queries to enhance the physical meaning of content queries in DETR. Moreover, we adopt a coarse-to-fine strategy to accurately identify graphical page objects. Experimental results demonstrate that our proposed DLAFormer outperforms previous approaches that employ multi-branch or multi-stage architectures for multiple tasks on two document layout analysis benchmarks, DocLayNet and Comp-HRDoc.
Discourse relation classification is an especially difficult task without explicit context markers (Prasad et al., 2008). Current approaches to implicit relation prediction solely rely on two neighboring sentences being targeted, ignoring the broader context of their surrounding environments (Atwell et al., 2021). In this research, we propose three new methods in which to incorporate context in the task of sentence relation prediction: (1) Direct Neighbors (DNs), (2) Expanded Window Neighbors (EWNs), and (3) Part-Smart Random Neighbors (PSRNs). Our findings indicate that the inclusion of context beyond one discourse unit is harmful in the task of discourse relation classification.
In the task of Knowledge Graph Completion (KGC), the existing datasets and their inherent subtasks carry a wealth of shared knowledge that can be utilized to enhance the representation of knowledge triplets and overall performance. However, no current studies specifically address the shared knowledge within KGC. To bridge this gap, we introduce a multi-level Shared Knowledge Guided learning method (SKG) that operates at both the dataset and task levels. On the dataset level, SKG-KGC broadens the original dataset by identifying shared features within entity sets via text summarization. On the task level, for the three typical KGC subtasks - head entity prediction, relation prediction, and tail entity prediction - we present an innovative multi-task learning architecture with dynamically adjusted loss weights. This approach allows the model to focus on more challenging and underperforming tasks, effectively mitigating the imbalance of knowledge sharing among subtasks. Experimental results demonstrate that SKG-KGC outperforms existing text-based methods significantly on three well-known datasets, with the most notable improvement on WN18RR.
Knowledge Graphs have been widely used to represent facts in a structured format. Due to their large scale applications, knowledge graphs suffer from being incomplete. The relation prediction task obtains knowledge graph completion by assigning one or more possible relations to each pair of nodes. In this work, we make use of the knowledge graph node names to fine-tune a large language model for the relation prediction task. By utilizing the node names only we enable our model to operate sufficiently in the inductive settings. Our experiments show that we accomplish new scores on a widely used knowledge graph benchmark.
Knowledge Graphs (KGs) are widely employed in artificial intelligence applications, such as question-answering and recommendation systems. However, KGs are frequently found to be incomplete. While much of the existing literature focuses on predicting missing nodes for given incomplete KG triples, there remains an opportunity to complete KGs by exploring relations between existing nodes, a task known as relation prediction. In this study, we propose a relations prediction model that harnesses both textual and structural information within KGs. Our approach integrates walks-based embeddings with language model embeddings to effectively represent nodes. We demonstrate that our model achieves competitive results in the relation prediction task when evaluated on a widely used dataset.
Information extraction (IE) is an important task in Natural Language Processing (NLP), involving the extraction of named entities and their relationships from unstructured text. In this paper, we propose a novel approach to this task by formulating it as graph structure learning (GSL). By formulating IE as GSL, we enhance the model's ability to dynamically refine and optimize the graph structure during the extraction process. This formulation allows for better interaction and structure-informed decisions for entity and relation prediction, in contrast to previous models that have separate or untied predictions for these tasks. When compared against state-of-the-art baselines on joint entity and relation extraction benchmarks, our model, GraphER, achieves competitive results.
Video Anomaly Detection (VAD), aiming to identify abnormalities within a specific context and timeframe, is crucial for intelligent Video Surveillance Systems. While recent deep learning-based VAD models have shown promising results by generating high-resolution frames, they often lack competence in preserving detailed spatial and temporal coherence in video frames. To tackle this issue, we propose a self-supervised learning approach for VAD through an inter-patch relationship prediction task. Specifically, we introduce a two-branch vision transformer network designed to capture deep visual features of video frames, addressing spatial and temporal dimensions responsible for modeling appearance and motion patterns, respectively. The inter-patch relationship in each dimension is decoupled into inter-patch similarity and the order information of each patch. To mitigate memory consumption, we convert the order information prediction task into a multi-label learning problem, and the inter-patch similarity prediction task into a distance matrix regression problem. Comprehensive experiments demonstrate the effectiveness of our method, surpassing pixel-generation-based methods by a significant margin across three public benchmarks. Additionally, our approach outperforms other self-supervised learning-based methods.
We introduce the Reality-linked 3D Scenes (R3DS) dataset of synthetic 3D scenes mirroring the real-world scene arrangements from Matterport3D panoramas. Compared to prior work, R3DS has more complete and densely populated scenes with objects linked to real-world observations in panoramas. R3DS also provides an object support hierarchy, and matching object sets (e.g., same chairs around a dining table) for each scene. Overall, R3DS contains 19K objects represented by 3,784 distinct CAD models from over 100 object categories. We demonstrate the effectiveness of R3DS on the Panoramic Scene Understanding task. We find that: 1) training on R3DS enables better generalization; 2) support relation prediction trained with R3DS improves performance compared to heuristically calculated support; and 3) R3DS offers a challenging benchmark for future work on panoramic scene understanding.
Spatial relationships between objects represent key scene information for humans to understand and interact with the world. To study the capability of current computer vision systems to recognize physically grounded spatial relations, we start by proposing precise relation definitions that permit consistently annotating a benchmark dataset. Despite the apparent simplicity of this task relative to others in the recognition literature, we observe that existing approaches perform poorly on this benchmark. We propose new approaches exploiting the long-range attention capabilities of transformers for this task, and evaluating key design principles. We identify a simple "RelatiViT" architecture and demonstrate that it outperforms all current approaches. To our knowledge, this is the first method to convincingly outperform naive baselines on spatial relation prediction in in-the-wild settings. The code and datasets are available in \url{https://sites.google.com/view/spatial-relation}.
The goal of few-shot relation extraction is to predict relations between name entities in a sentence when only a few labeled instances are available for training. Existing few-shot relation extraction methods focus on uni-modal information such as text only. This reduces performance when there are no clear contexts between the name entities described in text. We propose a multi-modal few-shot relation extraction model (MFS-HVE) that leverages both textual and visual semantic information to learn a multi-modal representation jointly. The MFS-HVE includes semantic feature extractors and multi-modal fusion components. The MFS-HVE semantic feature extractors are developed to extract both textual and visual features. The visual features include global image features and local object features within the image. The MFS-HVE multi-modal fusion unit integrates information from various modalities using image-guided attention, object-guided attention, and hybrid feature attention to fully capture the semantic interaction between visual regions of images and relevant texts. Extensive experiments conducted on two public datasets demonstrate that semantic visual information significantly improves the performance of few-shot relation prediction.
Existing methods for Visual Information Extraction (VIE) from form-like documents typically fragment the process into separate subtasks, such as key information extraction, key-value pair extraction, and choice group extraction. However, these approaches often overlook the hierarchical structure of form documents, including hierarchical key-value pairs and hierarchical choice groups. To address these limitations, we present a new perspective, reframing VIE as a relation prediction problem and unifying labels of different tasks into a single label space. This unified approach allows for the definition of various relation types and effectively tackles hierarchical relationships in form-like documents. In line with this perspective, we present UniVIE, a unified model that addresses the VIE problem comprehensively. UniVIE functions using a coarse-to-fine strategy. It initially generates tree proposals through a tree proposal network, which are subsequently refined into hierarchical trees by a relation decoder module. To enhance the relation prediction capabilities of UniVIE, we incorporate two novel tree constraints into the relation decoder: a tree attention mask and a tree level embedding. Extensive experimental evaluations on both our in-house dataset HierForms and a publicly available dataset SIBR, substantiate that our method achieves state-of-the-art results, underscoring the effectiveness and potential of our unified approach in advancing the field of VIE.
Abstract—The graph attention network (GAT) [1] has started to become a mainstream neural network architecture since 2018, yielding remarkable performance gains in various natural language processing (NLP) tasks. Although GAT has reached the state-of-the-art (SOTA) performance as a recent success in relation prediction in knowledge graph, the current model is still limited by the following two aspects: (1) the existing model only considers the neighbors from the inbound-direction of the given entity, but ignores the rich neighborhood information from outbounddirections; (2) the existing model only uses the k-th hop output to learn the multi-hop embeddings, which leads to the loss of a large amount of early-stage embedding information (e.g., one-hop) at the graph attention step. In this study, we propose a novel bidirectional graph attention network (BiGAT) to learn the hierarchical neighbor propagation. In our proposed BiGAT, an inbound-directional GAT and an outbound-directional GAT are introduced to capture sufficient neighborhood information before propagating the bidirectional neighborhood information to learn the multi-hop feature embeddings in a hierarchical manner. Experiments conducted on the four publicly available datasets show that BiGAT achieves the competitive results in comparison to other SOTA methods.
Aiming to accurately predict missing edges representing relations between entities, which are pervasive in real-world Knowledge Graphs (KGs), relation prediction plays a critical role in enhancing the comprehensiveness and utility of KGs. Recent research focuses on path-based methods due to their inductive and explainable properties. However, these methods face a great challenge when lots of reasoning paths do not form Closed Paths (CPs) in the KG. To address this challenge, we propose Anchoring Path Sentence Transformer (APST) by introducing Anchoring Paths (APs) to alleviate the reliance of CPs. Specifically, we develop a search-based description retrieval method to enrich entity descriptions and an assessment mechanism to evaluate the rationality of APs. APST takes both APs and CPs as the inputs of a unified Sentence Transformer architecture, enabling comprehensive predictions and high-quality explanations. We evaluate APST on three public datasets and achieve state-of-the-art (SOTA) performance in 30 of 36 transductive, inductive, and few-shot experimental settings.
This paper studies the problem of CPRP, concept prerequisite relation prediction, which is a fundamental task in using AI for education. CPRP is usually formulated into a link-prediction task on a relationship graph of concepts and solved by training the graph neural network (GNN) model. However, current directed GNNs fail to manage graph isomorphism which refers to the invariance of non-isomorphic graphs, reducing the expressivity of resulting representations. We present a permutation-equivariant directed GNN model by introducing the Weisfeiler-Lehman test into directed GNN learning. Our method is then used for CPRP and evaluated on three public datasets. The experimental results show that our model delivers better prediction performance than the state-of-the-art methods.
Panoptic Scene Graph Generation (PSG) aims at achieving a comprehensive image understanding by simultaneously segmenting objects and predicting relations among objects. However, the long-tail problem among relations leads to unsatisfactory results in real-world applications. Prior methods predominantly rely on vision information or utilize limited language information, such as object or relation names, thereby overlooking the utility of language information. Leveraging the recent progress in Large Language Models (LLMs), we propose to use language information to assist relation prediction, particularly for rare relations. To this end, we propose the Vision-Language Prompting (VLPrompt) model, which acquires vision information from images and language information from LLMs. Then, through a prompter network based on attention mechanism, it achieves precise relation prediction. Our extensive experiments show that VLPrompt significantly outperforms previous state-of-the-art methods on the PSG dataset, proving the effectiveness of incorporating language information and alleviating the long-tail problem of relations. Code is available at \url{https://github.com/franciszzj/TP-SIS}.
Most biomedical pretrained language models are monolingual and cannot handle the growing cross-lingual requirements. The scarcity of non-English domain corpora, not to mention parallel data, poses a significant hurdle in training multilingual biomedical models. Since knowledge forms the core of domain-specific corpora and can be translated into various languages accurately, we propose a model called KBioXLM, which transforms the multilingual pretrained model XLM-R into the biomedical domain using a knowledge-anchored approach. We achieve a biomedical multilingual corpus by incorporating three granularity knowledge alignments (entity, fact, and passage levels) into monolingual corpora. Then we design three corresponding training tasks (entity masking, relation masking, and passage relation prediction) and continue training on top of the XLM-R model to enhance its domain cross-lingual ability. To validate the effectiveness of our model, we translate the English benchmarks of multiple tasks into Chinese. Experimental results demonstrate that our model significantly outperforms monolingual and multilingual pretrained models in cross-lingual zero-shot and few-shot scenarios, achieving improvements of up to 10+ points. Our code is publicly available at https://github.com/ngwlh-gl/KBioXLM.
Expressing universal semantics common to all languages is helpful in understanding the meanings of complex and culture-specific sentences. The research theme underlying this scenario focuses on learning universal representations across languages with the usage of massive parallel corpora. However, due to the sparsity and scarcity of parallel data, there is still a big challenge in learning authentic ``universals'' for any two languages. In this paper, we propose EMMA-X: an EM-like Multilingual pre-training Algorithm, to learn (X)Cross-lingual universals with the aid of excessive multilingual non-parallel data. EMMA-X unifies the cross-lingual representation learning task and an extra semantic relation prediction task within an EM framework. Both the extra semantic classifier and the cross-lingual sentence encoder approximate the semantic relation of two sentences, and supervise each other until convergence. To evaluate EMMA-X, we conduct experiments on XRETE, a newly introduced benchmark containing 12 widely studied cross-lingual tasks that fully depend on sentence-level representations. Results reveal that EMMA-X achieves state-of-the-art performance. Further geometric analysis of the built representation space with three requirements demonstrates the superiority of EMMA-X over advanced models.
Scene graph generation (SGG) aims to automatically map an image into a semantic structural graph for better scene understanding. It has attracted significant attention for its ability to provide object and relation information, enabling graph reasoning for downstream tasks. However, it faces severe limitations in practice due to the biased data and training method. In this paper, we present a more rational and effective strategy based on causal inference for object relation prediction. To further evaluate the superiority of our strategy, we propose an object enhancement module to conduct ablation studies. Experimental results on the Visual Gnome 150 (VG-150) dataset demonstrate the effectiveness of our proposed method. These contributions can provide great potential for foundation models for decision-making.
Many downstream inference tasks for knowledge graphs, such as relation prediction, have been handled successfully by knowledge graph embedding techniques in the transductive setting. To address the inductive setting wherein new entities are introduced into the knowledge graph at inference time, more recent work opts for models which learn implicit representations of the knowledge graph through a complex function of a network's subgraph structure, often parametrized by graph neural network architectures. These come at the cost of increased parametrization, reduced interpretability and limited generalization to other downstream inference tasks. In this work, we bridge the gap between traditional transductive knowledge graph embedding approaches and more recent inductive relation prediction models by introducing a generalized form of harmonic extension which leverages representations learned through transductive embedding methods to infer representations of new entities introduced at inference time as in the inductive setting. This harmonic extension technique provides the best such approximation, can be implemented via an efficient iterative scheme, and can be employed to answer a family of conjunctive logical queries over the knowledge graph, further expanding the capabilities of transductive embedding methods. In experiments on a number of large-scale knowledge graph embedding benchmarks, we find that this approach for extending the functionality of transductive knowledge graph embedding models to perform knowledge graph completion and answer logical queries in the inductive setting is competitive with--and in some scenarios outperforms--several state-of-the-art models derived explicitly for such inductive tasks.