{"type":"ScholarlyArticle","title":"A Context-Enhanced Framework for Sequential Graph Reasoning","authors":[{"name":"Shuo Shi","slug":"shuo-shi","given":"Shuo","family":"Shi","affiliation":"Shanghai Key Laboratory of Trustworthy Computing, Software Engineering Institute , East China Normal University , Shanghai , China"},{"name":"Chao Peng","slug":"chao-peng","given":"Chao","family":"Peng","affiliation":"Shanghai Key Laboratory of Trustworthy Computing, Software Engineering Institute , East China Normal University , Shanghai , China"},{"name":"Chenyang Xu","slug":"chenyang-xu","given":"Chenyang","family":"Xu","affiliation":"Shanghai Key Laboratory of Trustworthy Computing, Software Engineering Institute , East China Normal University , Shanghai , China"},{"name":"Zhengfeng Yang","slug":"zhengfeng-yang","given":"Zhengfeng","family":"Yang","affiliation":"Shanghai Key Laboratory of Trustworthy Computing, Software Engineering Institute , East China Normal University , Shanghai , China"}],"abstract":"The paper studies sequential reasoning over graph-structured data, which stands as a fundamental task in various trending fields like automated math problem solving and neural graph algorithm learning, attracting a lot of research interest. Simultaneously managing both sequential and graph-structured information in such tasks presents a notable challenge. Over recent years, many neural architectures in the literature have emerged to tackle the issue. In this work, we generalize the existing architectures and propose a context-enhanced framework. The crucial innovation is that the reasoning of each step does not only rely on the outcome of the preceding step but also leverages the aggregation of information from more historical outcomes. The idea stems from our observation that in sequential graph reasoning, each step's outcome has a much stronger inner connection with each other compared to traditional seq-to-seq tasks. We show that the framework can effectively integrate with the existing methods, enhancing their reasoning abilities. Empirical evaluations are conducted on the challenging CLRS Reasoning Benchmark, and the results demonstrate that the proposed framework significantly improves the performance of existing architectures, yielding state-of-the-art results across the majority of the datasets within the benchmark.","doi":"10.24963/ijcai.2024/542","url":"https://ijcai.org/proceedings/2024/542","pdf":"https://ijcai.org/proceedings/2024/0542.pdf","markdown":"https://ijcai.org/proceedings/2024/542.md","bibtex":"https://ijcai.org/proceedings/2024/bibtex/542","volume":{"slug":"2024","title":"Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2024","url":"https://ijcai.org/proceedings/2024/","year":2024,"location":"Jeju, Korea","dates":"3-9 August 2024","editors":"Kate Larson","isbn":"978-1-956792-04-1","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2024"},"section":"Machine Learning","track":"Main Track","number":542,"pages":{"first":4902,"last":4910},"keywords":["Machine Learning: ML: Sequence and graph learning"],"videos":[],"references":[{"position":1,"raw":"[Cormen et al., 2009] Thomas H. Cormen, Charles E. Leis- erson, Ronald L. Rivest, and Clifford Stein. Introduction to Algorithms, 3rd Edition. MIT Press, 2009.","title":"Introduction to Algorithms","authors":"H Thomas, Charles E Cormen, Ronald L Leiserson, Clifford Rivest, Stein","year":"2009","venue":null,"doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[de Rezende et al., 2021] Susanna F. de Rezende, Massimo Lauria, Jakob Nordström, and Dmitry Sokolov. The power of negative reasoning. In CCC, volume 200 of LIPIcs, pages 40:1-40:24. Schloss Dagstuhl -Leibniz-Zentrum für Informatik, 2021.","title":"The power of negative reasoning","authors":"Susanna F De Rezende, Massimo Lauria, Jakob Nordström, Dmitry Sokolov","year":"2021","venue":"CCC","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Diao and Loynd, 2023] Cameron Diao and Ricky Loynd. Relational attention: Generalizing transformers for graph- structured tasks. In ICLR. OpenReview.net, 2023.","title":"Relational attention: Generalizing transformers for graphstructured tasks","authors":"Cameron Diao, Ricky Loynd","year":"2023","venue":"ICLR. OpenReview.net","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Drori et al., 2020] Iddo Drori, Anant Kharkar, William R. Sickinger, Brandon Kates, Qiang Ma, Suwen Ge, Eden Dolev, Brenda Dietrich, David P. Williamson, and Madeleine Udell. Learning to solve combinatorial opti- mization problems on real-world graphs in linear time. In ICMLA, pages 19-24. IEEE, 2020.","title":"Learning to solve combinatorial optimization problems on real-world graphs in linear time","authors":"Iddo Drori, Anant Kharkar, William R Sickinger, Brandon Kates, Qiang Ma, Suwen Ge, Eden Dolev, Brenda Dietrich, David P Williamson, Madeleine Udell","year":"2020","venue":"ICMLA","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Garland and Gregg, 2018] James Garland and David Gregg. Low complexity multiply-accumulate units for convolu- tional neural networks with weight-sharing. ACM Trans. Archit. Code Optim., 15(3):31:1-31:24, 2018.","title":"Low complexity multiply-accumulate units for convolutional neural networks with weight-sharing","authors":"Gregg ; Garland, James Garland, David Gregg","year":"2018","venue":"ACM Trans. Archit. Code Optim","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Gerasimova et al., 2023] Olga Gerasimova, Nikita Severin, and Ilya Makarov. Comparative analysis of logic rea- soning and graph neural networks for ontology-mediated query answering with a covering axiom. IEEE Access, 11:88074-88086, 2023.","title":"Comparative analysis of logic reasoning and graph neural networks for ontology-mediated query answering with a covering axiom","authors":"Olga Gerasimova, Nikita Severin, Ilya Makarov","year":"2023","venue":"IEEE Access","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Gilmer et al., 2017] Justin Gilmer, Samuel S. Schoenholz, Patrick F. Riley, Oriol Vinyals, and George E. Dahl. Neu- ral message passing for quantum chemistry. In ICML, vol- ume 70 of Proceedings of Machine Learning Research, pages 1263-1272. PMLR, 2017.","title":"Neural message passing for quantum chemistry","authors":"Justin Gilmer, Samuel S Schoenholz, Patrick F Riley, Oriol Vinyals, George E Dahl","year":"2017","venue":"Proceedings of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[He et al., 2023] Weinan He, Canming Huang, Zhanhao Xiao, and Yongmei Liu. Exploring the capacity of pre- trained language models for reasoning about actions and change. In ACL (1), pages 4629-4643. Association for Computational Linguistics, 2023.","title":"Exploring the capacity of pretrained language models for reasoning about actions and change","authors":"Weinan He, Canming Huang, Zhanhao Xiao, Yongmei Liu","year":"2023","venue":"ACL (1)","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Huang et al., 2022] Zeren Huang, Kerong Wang, Furui Liu, Hui-Ling Zhen, Weinan Zhang, Mingxuan Yuan, Jianye Hao, Yong Yu, and Jun Wang. Learning to select cuts for efficient mixed-integer programming. Pattern Recognit., 123:108353, 2022.","title":"Learning to select cuts for efficient mixed-integer programming","authors":"Zeren Huang, Kerong Wang, Furui Liu, Hui-Ling Zhen, Weinan Zhang, Mingxuan Yuan, Jianye Hao, Yong Yu, Jun Wang","year":"2022","venue":"Pattern Recognit","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Hussain et al., 2022] Md. Shamim Hussain, Mohammed J. Zaki, and Dharmashankar Subramanian. Global self- attention as a replacement for graph convolution. In KDD, pages 655-665. ACM, 2022.","title":"Global selfattention as a replacement for graph convolution","authors":"Shamim Md, Mohammed J Hussain, Dharmashankar Zaki, Subramanian","year":"2022","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Ibarz et al., 2022] Borja Ibarz, Vitaly Kurin, George Papa- makarios, Kyriacos Nikiforou, Mehdi Bennani, Róbert Csordás, Andrew Joseph Dudzik, Matko Bosnjak, Alex Vitvitskyi, Yulia Rubanova, Andreea Deac, Beatrice Bevilacqua, Yaroslav Ganin, Charles Blundell, and Petar Velickovic. A generalist neural algorithmic learner. In LoG, volume 198 of Proceedings of Machine Learning Re- search, page 2. PMLR, 2022.","title":"A generalist neural algorithmic learner","authors":"Borja Ibarz, Vitaly Kurin, George Papamakarios, Kyriacos Nikiforou, Mehdi Bennani, Róbert Csordás, Andrew Joseph Dudzik, Matko Bosnjak, Alex Vitvitskyi, Yulia Rubanova, Andreea Deac, Beatrice Bevilacqua, Yaroslav Ganin, Charles Blundell, Petar Velickovic","year":"2022","venue":"Proceedings of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Jain et al., 2023] Rishabh Jain, Petar Velickovic, and Pietro Liò. Neural priority queues for graph neural networks. In The 2023 ICML, Workshop on Knowledge and Logi- cal Reasoning in the Era of Data-driven Learning, volume 202. PMLR, 2023.","title":"Neural priority queues for graph neural networks","authors":"Rishabh Jain, Petar Velickovic, Pietro Liò","year":"2023","venue":"The 2023 ICML, Workshop on Knowledge and Logical Reasoning in the Era of Data-driven Learning","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Jiang et al., 2022] Jinhao Jiang, Kun Zhou, Ji-Rong Wen, and Wayne Xin Zhao. $great truths are always simple: $ A rather simple knowledge encoder for enhancing the commonsense reasoning capacity of pre-trained models. In NAACL-HLT (Findings), pages 1730-1741. Association for Computational Linguistics, 2022.","title":"$great truths are always simple: $ A rather simple knowledge encoder for enhancing the commonsense reasoning capacity of pre-trained models","authors":"Jinhao Jiang, Kun Zhou, Ji-Rong Wen, Wayne Xin Zhao","year":"2022","venue":"NAACL-HLT (Findings)","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Jürß et al., 2023] Jonas Jürß, Dulhan Hansaja Jayalath, and Petar Veličković. Recursive algorithmic reasoning. In The Second Learning on Graphs Conference, 2023.","title":"Recursive algorithmic reasoning","authors":"Jürß","year":"2023","venue":"The Second Learning on Graphs Conference","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Kipf and Welling, 2017] Thomas N. Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. In ICLR (Poster). OpenReview.net, 2017.","title":"Semi-supervised classification with graph convolutional networks","authors":"Thomas N Kipf, Max Welling","year":"2017","venue":"ICLR (Poster). OpenReview.net","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Lample and Charton, 2020] Guillaume Lample and Franc ¸ois Charton. Deep learning for symbolic mathemat- ics. In ICLR. OpenReview.net, 2020.","title":"Guillaume Lample and Franc ¸ois Charton. Deep learning for symbolic mathematics","authors":"Charton Lample","year":"2020","venue":"ICLR. OpenReview.net","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Li et al., 2023] Hongyang Li, Cui Zhu, Wenjun Zhu, Yun- feng Zhou, and Xinyu Zhang. Short-term sequential pat- tern for temporal knowledge graph reasoning. In CAIBDA, volume 373 of Frontiers in Artificial Intelligence and Ap- plications, pages 728-739. IOS Press, 2023.","title":"Short-term sequential pattern for temporal knowledge graph reasoning","authors":"Hongyang Li, Cui Zhu, Wenjun Zhu, Yunfeng Zhou, Xinyu Zhang","year":"2023","venue":"Frontiers in Artificial Intelligence and Applications","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Ling et al., 2022] Chen Ling, Tanmoy Chowdhury, Junji Jiang, Junxiang Wang, Xuchao Zhang, Haifeng Chen, and Liang Zhao. Deepgar: Deep graph learning for analogical reasoning. In ICDM, pages 1065-1070. IEEE, 2022.","title":"Deepgar: Deep graph learning for analogical reasoning","authors":"Chen Ling, Tanmoy Chowdhury, Junji Jiang, Junxiang Wang, Xuchao Zhang, Haifeng Chen, Liang Zhao","year":"2022","venue":"ICDM","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Ma et al., 2023] Ruixin Ma, Biao Mei, Yunlong Ma, Hongyan Zhang, Meihong Liu, and Liang Zhao. One- shot relational learning for extrapolation reasoning on temporal knowledge graphs. Data Min. Knowl. Discov., 37(4):1591-1608, 2023.","title":"Oneshot relational learning for extrapolation reasoning on temporal knowledge graphs","authors":"Ruixin Ma, Biao Mei, Yunlong Ma, Hongyan Zhang, Meihong Liu, Liang Zhao","year":"2023","venue":"Data Min. Knowl. Discov","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Numeroso et al., 2023] Danilo Numeroso, Davide Bacciu, and Petar Velickovic. Dual algorithmic reasoning. In ICLR. OpenReview.net, 2023.","title":"Dual algorithmic reasoning","authors":"Danilo Numeroso, Davide Bacciu, Petar Velickovic","year":"2023","venue":"ICLR. OpenReview.net","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Prim, 1957] R. C. Prim. Shortest connection networks and some generalizations. The Bell System Technical Journal, 36(6):1389-1401, 1957.","title":"Shortest connection networks and some generalizations","authors":"R C Prim","year":"1957","venue":"The Bell System Technical Journal","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Ramesh et al., 2023] Gowtham Ramesh, Makesh Nar- simhan Sreedhar, and Junjie Hu. Single sequence prediction over reasoning graphs for multi-hop QA. In ACL (1), pages 11466-11481. Association for Computa- tional Linguistics, 2023.","title":"Single sequence prediction over reasoning graphs for multi-hop QA","authors":"Gowtham Ramesh, Makesh Narsimhan Sreedhar, Junjie Hu","year":"2023","venue":"ACL (1)","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Schreck and Mathis, 2016] Pascal Schreck and Pascal Mathis. Automatic constructibility checking of a corpus of geometric construction problems. Math. Comput. Sci., 10(1):41-56, 2016.","title":"Automatic constructibility checking of a corpus of geometric construction problems","authors":"Mathis Schreck, Pascal Schreck, Pascal Mathis","year":"2016","venue":"Math. Comput. Sci","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Sukhbaatar et al., 2015] Sainbayar Sukhbaatar, Arthur Szlam, Jason Weston, and Rob Fergus. End-to-end memory networks. In NIPS, pages 2440-2448, 2015.","title":"End-to-end memory networks","authors":"Sukhbaatar","year":"2015","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Vaswani et al., 2017] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. In NIPS, pages 5998-6008, 2017.","title":"Attention is all you need","authors":"Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Lukasz Kaiser, Illia Polosukhin","year":"2017","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Velickovic et al., 2018] Petar Velickovic, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, and Yoshua Bengio. Graph attention networks. In ICLR (Poster). OpenReview.net, 2018.","title":"Graph attention networks","authors":"Petar Velickovic, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, Yoshua Bengio","year":"2018","venue":"ICLR (Poster). OpenReview.net","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Velickovic et al., 2020] Petar Velickovic, Lars Buesing, Matthew C. Overlan, Razvan Pascanu, Oriol Vinyals, and Charles Blundell. Pointer graph networks. In NeurIPS, 2020.","title":"Pointer graph networks","authors":"Petar Velickovic, Lars Buesing, Matthew C Overlan, Razvan Pascanu, Oriol Vinyals, Charles Blundell","year":"2020","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Velickovic et al., 2022] Petar Velickovic, Adrià Puig- domènech Badia, David Budden, Razvan Pascanu, Andrea Banino, Misha Dashevskiy, Raia Hadsell, and Charles Blundell. The CLRS algorithmic reasoning benchmark. In ICML, volume 162 of Proceedings of Machine Learning Research, pages 22084-22102. PMLR, 2022.","title":"The CLRS algorithmic reasoning benchmark","authors":"Petar Velickovic, Adrià Puigdomènech Badia, David Budden, Razvan Pascanu, Andrea Banino, Misha Dashevskiy, Raia Hadsell, Charles Blundell","year":"2022","venue":"Proceedings of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Wang et al., 2023] Yuan Wang, Kun Yu, Chen Chen, Xiyuan Hu, and Silong Peng. Dynamic graph learning with content-guided spatial-frequency relation reasoning for deepfake detection. In CVPR, pages 7278-7287. IEEE, 2023.","title":"Dynamic graph learning with content-guided spatial-frequency relation reasoning for deepfake detection","authors":"Yuan Wang, Kun Yu, Chen Chen, Xiyuan Hu, Silong Peng","year":"2023","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Wu et al., 2019] Jun Wu, Jingrui He, and Jiejun Xu. Demo- net: Degree-specific graph neural networks for node and graph classification. In KDD, pages 406-415. ACM, 2019.","title":"Demonet: Degree-specific graph neural networks for node and graph classification","authors":"Wu","year":"2019","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Wu et al., 2021] Zhanghao Wu, Paras Jain, Matthew A. Wright, Azalia Mirhoseini, Joseph E. Gonzalez, and Ion Stoica. Representing long-range context for graph neural networks with global attention. In NeurIPS, pages 13266- 13279, 2021.","title":"Representing long-range context for graph neural networks with global attention","authors":"Zhanghao Wu, Paras Jain, Matthew A Wright, Azalia Mirhoseini, Joseph E Gonzalez, Ion Stoica","year":"2021","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Yu, 2017] Lei Yu. Tackling sequence to sequence mapping problems with neural networks. PhD thesis, University of Oxford, UK, 2017.","title":"Tackling sequence to sequence mapping problems with neural networks","authors":"Lei Yu","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":33,"raw":"[Zhao et al., 2024] Zhenbing Zhao, Bin Liu, Yongjie Zhai, Wenqing Zhao, and Pan Su. Dual graph reasoning net- work for oil leakage segmentation in substation equip- ment. IEEE Trans. Instrum. Meas., 73:1-15, 2024.","title":"Dual graph reasoning network for oil leakage segmentation in substation equipment","authors":"Zhao","year":"2024","venue":"IEEE Trans. Instrum. Meas","doi":null,"url":null,"internal_url":null},{"position":34,"raw":"[Zhu, 2022] Zihao Zhu. From shallow to deep: Composi- tional reasoning over graphs for visual question answering. In ICASSP, pages 8217-8221. IEEE, 2022.","title":"From shallow to deep: Compositional reasoning over graphs for visual question answering","authors":"Zihao Zhu, Zhu","year":"2022","venue":"ICASSP","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}