Challenge: Existing methods emphasize contextual semantics while others pay more attention to explicit logical features. Existing models utilize graph convolutional networks (GCN) for node updates, still exhibiting some shortcomings.
Approach: They propose a logical reasoning method with contrastive learning and lightweight graph networks (LogiGraph) they employ conjunction and punctuation marks as two types of edges to construct a dual graph.
Outcome: The proposed method improves the GCN and employs conjunction and punctuation marks as two types of edges to construct a dual graph.

Similar Papers

AdaLoGN: Adaptive Logic Graph Network for Reasoning-Based Machine Reading Comprehension (2022.acl-long)

Copied to clipboard

Challenge: Existing methods and limitations for machine reading comprehension are insufficient for logical reasoning over text.
Approach: They propose a neural-symbolic approach which passes messages over a graph representing logical relations between text units to predict an answer.
Outcome: The proposed approach outperforms existing methods on ReClor and LogiQA.
DAGN: Discourse-Aware Graph Network for Logical Reasoning (2021.naacl-main)

Copied to clipboard

Challenge: Recent QA with logical reasoning questions requires passage-level relations among the sentences.
Approach: They propose a discourse-aware graph network that aggregates passage-level clues for QA by using discourse-based information.
Outcome: The proposed model achieves competitive results on two logical reasoning QA datasets.
Inductive Relation Prediction with Logical Reasoning Using Contrastive Representations (2022.emnlp-main)

Copied to clipboard

Challenge: Existing methods for relation prediction in knowledge graphs (KGs) are limited by the inductive setting because entities in training process are finite.
Approach: They propose a graph convolutional network-based model LogCo with logical reasoning by contrastive representations that extracts subgraphs and relational paths between two entities to supply the entity-independence.
Outcome: The proposed model outperforms existing methods on twelve inductive datasets.
Be More with Less: Hypergraph Attention Networks for Inductive Text Classification (2020.emnlp-main)

Copied to clipboard

Challenge: Text classification is a critical research topic with broad applications in natural language processing. graph neural networks (GNNs) have received increasing attention but their performance is jeopardized in practice.
Approach: They propose a model which captures long-distance interactions between words and a graph-based model which can be used to perform text classification.
Outcome: The proposed model can achieve more expressive power with less computational consumption on the text classification task.
Coarse-to-Fine Contrastive Learning in Image-Text-Graph Space for Improved Vision-Language Compositionality (2023.emnlp-main)

Copied to clipboard

Challenge: Recent studies have highlighted severe limitations of contrastive learning models in their ability to perform compositional reasoning over objects, attributes, and relations.
Approach: They propose a graph decomposition framework and negative mining techniques to improve attribute binding and relation understanding of scene graphs.
Outcome: The proposed approach improves attribute binding, relation understanding, generalization, and productivity on multiple benchmarks.
Doc-GCN: Heterogeneous Graph Convolutional Networks for Document Layout Analysis (2022.coling-1)

Copied to clipboard

Challenge: Document Layout Analysis tasks rely on visual cues to understand documents . traditional deep learning-based methods fail to recognize the layout and components of unstructured documents based on the document structure and the boundaries of each layout region.
Approach: They propose a way to harmonize and integrate heterogeneous aspects for Document Layout Analysis by using graph convolutional networks to enhance each aspect of features.
Outcome: The proposed task is based on three widely used datasets: PubLayNet, FUNSD, and DocBank.
Contrastive Self-Supervised Learning for Commonsense Reasoning (2020.acl-main)

Copied to clipboard

Challenge: Existing methods for commonsense reasoning are limited by current methods . empirical results show that our method alleviates the limitation of current supervised approaches .
Approach: They propose a self-supervised method to solve pronoun disambiguation problems . they leverage a mutual exclusive loss regularized by a contrastive margin to achieve commonsense reasoning .
Outcome: The proposed method performs well on many NLP benchmarks.
C3LPGCN:Integrating Contrastive Learning and Cooperative Learning with Prompt into Graph Convolutional Network for Aspect-based Sentiment Analysis (2024.findings-naacl)

Copied to clipboard

Challenge: Recent studies have shown that graph convolutional networks (GCNs) can model syntactic information but incorrect syntaktic structure may introduce additional noise.
Approach: They propose a graph convolutional network which integrates Contrastive Learning and Cooperative Learning with Prompt into GCN to alleviate the noise when modeling syntactic information.
Outcome: The proposed model outperforms state-of-the-art models on three datasets and significantly outperformed existing models.
Question Answering by Reasoning Across Documents with Graph Convolutional Networks (N19-1)

Copied to clipboard

Challenge: Recent research in reading comprehension has focused on answering questions based on individual documents or even single paragraphs.
Approach: They propose a neural model which integrates and reasons relying on information spread within documents and across multiple documents.
Outcome: The proposed model achieves state-of-the-art on a multi-document question answering dataset, WikiHop.
Question Directed Graph Attention Network for Numerical Reasoning over Text (2020.emnlp-main)

Copied to clipboard

Challenge: Numerical reasoning requires both natural language understanding and arithmetic computation.
Approach: They propose a graph representation for the context of the passage and question needed for numerical reasoning.
Outcome: The proposed model achieves remarkable results in benchmark datasets such as DROP.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations