Harry Shomer
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Harry Shomer has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Jiliang Tang
Michigan State University, East Lansing, MI, USA
6 shared publications - Hui Liu
Michigan State University, East Lansing, MI, USA
5 shared publications - Haoyu Han
Michigan State University, East Lansing, USA
3 shared publications - Yu Wang
University of Oregon
3 shared publications - Bo Long
Monetization AI, Meta, Menlo Park, CA, USA
2 shared publications - Jay Revolinsky
Michigan State University, East Lansing, MI, USA
2 shared publications
Latest Publications
- Top Cited PapersOpen Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 8966-8977
RAG vs. GraphRAG: A Systematic Evaluation and Key Insights
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 8966-8977
Top Cited PapersOpen AccessRAG vs. GraphRAG: A Systematic Evaluation and Key Insights
- Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1367-1370
Rigorizing Retrieval-augmented Generation with Structured Knowledge Intelligence (6 Hrs)
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1367-1370
Open AccessRigorizing Retrieval-augmented Generation with Structured Knowledge Intelligence (6 Hrs)
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 6296-6297
Machine Learning on Graphs in the Era of Generative Artificial Intelligence
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 6296-6297
Open AccessMachine Learning on Graphs in the Era of Generative Artificial Intelligence
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5777-5788
Towards Better Benchmark Datasets for Inductive Knowledge Graph Completion
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5777-5788
Open AccessTowards Better Benchmark Datasets for Inductive Knowledge Graph Completion
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5766-5776
Towards Understanding Link Predictor Generalizability Under Distribution Shifts
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5766-5776
Open AccessTowards Understanding Link Predictor Generalizability Under Distribution Shifts
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 2621-2632
A Scalable Pretraining Framework for Link Prediction with Efficient Adaptation
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 2621-2632
Open AccessA Scalable Pretraining Framework for Link Prediction with Efficient Adaptation
- Open Access
- 24 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 2686-2698
LPFormer: An Adaptive Graph Transformer for Link Prediction
Link prediction is a common task on graph-structured data that has seen applications in a variety of domains. Classically, hand-crafted heuristics were used for this task. Heuristic measures are chosen such that they correlate well with the underlying factors related to link formation. In recent years, a new class of methods has emerged that combines the advantages of message-passing neural networks (MPNN) and heuristics methods. These methods perform predictions by using the output of an MPNN in conjunction with a "pairwise encoding" that captures the relationship between nodes in the candidate link. They have been shown to achieve strong performance on numerous datasets. However, current pairwise encodings often contain a strong inductive bias, using the same underlying factors to classify all links. This limits the ability of existing methods to learn how to properly classify a variety of different links that may form from different factors. To address this limitation, we propose a new method, LPFormer, which attempts to adaptively learn the pairwise encodings for each link. LPFormer models the link factors via an attention module that learns the pairwise encoding that exists between nodes by modeling multiple factors integral to link prediction. Extensive experiments demonstrate that LPFormer can achieve SOTA performance on numerous datasets while maintaining efficiency. The code is available at The code is available at https://github.com/HarryShomer/LPFormer.
- 24 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 2686-2698
Open AccessLPFormer: An Adaptive Graph Transformer for Link Prediction
Link prediction is a common task on graph-structured data that has seen applications in a variety of domains. Classically, hand-crafted heuristics were used for this task. Heuristic measures are chosen such that they correlate well with the underlying factors related to link formation. In recent years, a new class of methods has emerged that combines the advantages of message-passing neural networks (MPNN) and heuristics methods. These methods perform predictions by using the output of an MPNN in conjunction with a "pairwise encoding" that captures the relationship between nodes in the candidate link. They have been shown to achieve strong performance on numerous datasets. However, current pairwise encodings often contain a strong inductive bias, using the same underlying factors to classify all links. This limits the ability of existing methods to learn how to properly classify a variety of different links that may form from different factors. To address this limitation, we propose a new method, LPFormer, which attempts to adaptively learn the pairwise encodings for each link. LPFormer models the link factors via an attention module that learns the pairwise encoding that exists between nodes by modeling multiple factors integral to link prediction. Extensive experiments demonstrate that LPFormer can achieve SOTA performance on numerous datasets while maintaining efficiency. The code is available at The code is available at https://github.com/HarryShomer/LPFormer.
- Open Access
- 6 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 253-257
Learning Representations for Hyper-Relational Knowledge Graphs
Knowledge graphs (KGs) have gained prominence for their ability to learn representations for uni-relational facts. Recently, research has focused on modeling hyper-relational facts, which move beyond the restriction of uni-relational facts and allow us to represent more complex and real-world information. However, existing approaches for learning representations on hyper-relational KGs majorly focus on enhancing the communication from qualifiers to base triples while overlooking the flow of information from base triple to qualifiers. This can lead to suboptimal qualifier representations, especially when a large amount of qualifiers are presented. It motivates us to design a framework that utilizes multiple aggregators to learn representations for hyper-relational facts: one from the perspective of the base triple and the other one from the perspective of the qualifiers. Experiments demonstrate the effectiveness of our framework for hyper-relational knowledge graph completion across multiple datasets. Furthermore, we conduct an ablation study that validates the importance of the various components in our framework.
- 6 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 253-257
Open AccessLearning Representations for Hyper-Relational Knowledge Graphs
Knowledge graphs (KGs) have gained prominence for their ability to learn representations for uni-relational facts. Recently, research has focused on modeling hyper-relational facts, which move beyond the restriction of uni-relational facts and allow us to represent more complex and real-world information. However, existing approaches for learning representations on hyper-relational KGs majorly focus on enhancing the communication from qualifiers to base triples while overlooking the flow of information from base triple to qualifiers. This can lead to suboptimal qualifier representations, especially when a large amount of qualifiers are presented. It motivates us to design a framework that utilizes multiple aggregators to learn representations for hyper-relational facts: one from the perspective of the base triple and the other one from the perspective of the qualifiers. Experiments demonstrate the effectiveness of our framework for hyper-relational knowledge graph completion across multiple datasets. Furthermore, we conduct an ablation study that validates the importance of the various components in our framework.
- Open Access
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 2637-2645
A Mix-up Strategy to Enhance Adversarial Training with Imbalanced Data
Adversarial training has been proven to be one of the most effective techniques to defend against adversarial examples. The majority of existing adversarial training methods assume that every class in the training data is equally distributed. However, in reality, some classes often have a large number of training data while others only have a very limited amount. Recent studies have shown that the performance of adversarial training will degrade drastically if the training data is imbalanced. In this paper, we propose a simple yet effective framework to enhance the robustness of DNN models under imbalanced scenarios. Our framework, Imb-Mix, first augments the training dataset by generating multiple adversarial examples for samples in the minority classes. This is done by first adding random noise to the original adversarial examples created by one specific adversarial attack method. It then constructs Mixup-mimic mixed examples upon the augmented dataset used by adversarial training. In addition, we theoretically prove the regularization effect of our Mixup-mimic mixed examples generation technique in Imb-Mix. Extensive experiments on various imbalanced datasets verify the effectiveness of the proposed framework.
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 2637-2645
Open AccessA Mix-up Strategy to Enhance Adversarial Training with Imbalanced Data
Adversarial training has been proven to be one of the most effective techniques to defend against adversarial examples. The majority of existing adversarial training methods assume that every class in the training data is equally distributed. However, in reality, some classes often have a large number of training data while others only have a very limited amount. Recent studies have shown that the performance of adversarial training will degrade drastically if the training data is imbalanced. In this paper, we propose a simple yet effective framework to enhance the robustness of DNN models under imbalanced scenarios. Our framework, Imb-Mix, first augments the training dataset by generating multiple adversarial examples for samples in the minority classes. This is done by first adding random noise to the original adversarial examples created by one specific adversarial attack method. It then constructs Mixup-mimic mixed examples upon the augmented dataset used by adversarial training. In addition, we theoretically prove the regularization effect of our Mixup-mimic mixed examples generation technique in Imb-Mix. Extensive experiments on various imbalanced datasets verify the effectiveness of the proposed framework.
- Open Access
- 30 April 2023
- Published by Association for Computing Machinery (ACM)
- p. 705-715
Toward Degree Bias in Embedding-Based Knowledge Graph Completion
A fundamental task for knowledge graphs (KGs) is knowledge graph completion (KGC). It aims to predict unseen edges by learning representations for all the entities and relations in a KG. A common concern when learning representations on traditional graphs is degree bias. It can affect graph algorithms by learning poor representations for lower-degree nodes, often leading to low performance on such nodes. However, there has been limited research on whether there exists degree bias for embedding-based KGC and how such bias affects the performance of KGC. In this paper, we validate the existence of degree bias in embedding-based KGC and identify the key factor to degree bias. We then introduce a novel data augmentation method, KG-Mixup, to generate synthetic triples to mitigate such bias. Extensive experiments have demonstrated that our method can improve various embedding-based KGC methods and outperform other methods tackling the bias problem on multiple benchmark datasets. 1
- 30 April 2023
- Published by Association for Computing Machinery (ACM)
- p. 705-715
Open AccessToward Degree Bias in Embedding-Based Knowledge Graph Completion
A fundamental task for knowledge graphs (KGs) is knowledge graph completion (KGC). It aims to predict unseen edges by learning representations for all the entities and relations in a KG. A common concern when learning representations on traditional graphs is degree bias. It can affect graph algorithms by learning poor representations for lower-degree nodes, often leading to low performance on such nodes. However, there has been limited research on whether there exists degree bias for embedding-based KGC and how such bias affects the performance of KGC. In this paper, we validate the existence of degree bias in embedding-based KGC and identify the key factor to degree bias. We then introduce a novel data augmentation method, KG-Mixup, to generate synthetic triples to mitigate such bias. Extensive experiments have demonstrated that our method can improve various embedding-based KGC methods and outperform other methods tackling the bias problem on multiple benchmark datasets. 1