pith. sign in

arxiv: 2606.18621 · v1 · pith:MHLZJ7PInew · submitted 2026-06-17 · 💻 cs.LG

Towards Anomaly Detection on Relational Data

Shiyuan Li , Yunfeng Zhao , Yue Tan , Qingfeng Chen , Yixin Liu , Shirui Pan This is my paper

Pith reviewed 2026-06-26 21:57 UTC · model grok-4.3

classification 💻 cs.LG
keywords anomaly detectionrelational databasesreconstruction-basedmulti-table attributesforeign-key relationsgraph anomaly detectionattribute reconstructionedge reconstruction
0
0 comments X

The pith

RelAD detects anomalies in relational databases by reconstructing both multi-table attributes and foreign-key connections.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes RelAD as a reconstruction-based method to find anomalies in relational data that spans multiple tables. It targets two main difficulties: high-dimensional heterogeneous attributes across tables that can hide sparse abnormal signals, and abnormal patterns in foreign-key links between entities. The approach rebuilds attributes with a conditional sparse-gated module to focus on relevant semantic blocks while ignoring noise, and it rebuilds edges from two views of entity profiles to spot relation-specific irregularities. These two signals combine in a fusion step to yield an anomaly score. If the claim holds, it would allow reliable anomaly detection in common real-world systems such as fraud monitoring or risk assessment where existing tabular and graph methods fall short.

Core claim

RelAD is a reconstruction-based framework that captures anomalies from both attribute and relational edge reconstruction. It contains conditional sparse-gated attribute reconstruction, which suppresses redundant multi-table attributes and emphasizes abnormal semantic blocks, and dual-view multi-relational edge reconstruction, which detects relation-specific abnormal connections from both intrinsic and behavioral entity profiles. The resulting attribute and relational signals are integrated through a lightweight fusion module to produce the final anomaly score. On six constructed benchmark datasets with systematic anomalies, RelAD consistently outperforms other baselines while achieving compe

What carries the argument

Dual reconstruction: conditional sparse-gated attribute reconstruction combined with dual-view multi-relational edge reconstruction, fused for the anomaly score.

If this is right

  • Anomaly detection becomes feasible for relational data where abnormal clues are sparse across many heterogeneous attributes.
  • Detection can now target abnormal foreign-key connection patterns that standard tabular and graph methods overlook.
  • A single framework can produce competitive accuracy and runtime on multiple relational benchmarks.
  • The method supplies separate attribute-level and relation-level signals that can be inspected for explanation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same reconstruction signals might be used to rank which specific attributes or links are most anomalous in a given case.
  • The benchmark construction process could be reused to create additional test sets for other relational anomaly settings.
  • The fusion module might be replaced by more advanced integration techniques without changing the core reconstruction modules.

Load-bearing premise

The six constructed benchmark datasets with systematic anomalies sufficiently capture the real-world challenges of high-dimensional heterogeneous multi-table attributes and abnormal foreign-key connection patterns.

What would settle it

An evaluation on real production relational databases that contain verified anomalies in which RelAD does not outperform standard tabular or graph anomaly detection baselines.

Figures

Figures reproduced from arXiv: 2606.18621 by Qingfeng Chen, Shirui Pan, Shiyuan Li, Yixin Liu, Yue Tan, Yunfeng Zhao.

Figure 1
Figure 1. Figure 1: Sketch maps of (a) Anomalous entity in relational database; and the limitations of adapting [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The overall pipeline of RelAD for relational data anomaly detection. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Sensitivity w.r.t. α and β. For β, most datasets prefer moderate values, confirm￾ing that both relational views provide complementary ev￾idence. This disparity suggests that anomalies in different relational databases may manifest primarily in attributes or in relational connections, and RelAD can flexibly adapt the fusion weights to the underlying anomaly source. Efficiency Analysis. To assess the runtime… view at source ↗
Figure 4
Figure 4. Figure 4: Runtime comparison on the HM dataset. In this section, we briefly review three lines of related studies. A detailed literature review is in Appendix G. Relational Deep Learning (RDL) aims to learn from multi-table relational databases by model￾ing entities and their primary–foreign key rela￾tionships [27]. Mainstream studies transform relational databases into heterogeneous graphs and develop graph neural … view at source ↗
Figure 5
Figure 5. Figure 5: Sensitivity analysis of RelAD with respect to [PITH_FULL_IMAGE:figures/full_fig_p021_5.png] view at source ↗
read the original abstract

Relational databases are widely used for managing structured data in real-world systems. Detecting anomalies from such relational data is crucial for identifying fraud, risks, and abnormal behaviors, yet remains under-explored. The key challenges lie in the intrinsic complexity of relational data: multi-table attributes are high-dimensional and heterogeneous, making sparse abnormal clues easy to overwhelm by normal or irrelevant information; and anomalies may further manifest as abnormal connection patterns across different foreign-key relations, which existing tabular and graph anomaly detection methods are ill-suited to capture. To address them, we propose RelAD, a reconstruction-based framework that captures anomalies from both attribute and relational edge reconstruction. RelAD contains two core modules: conditional sparse-gated attribute reconstruction, which suppresses redundant multi-table attributes and emphasizes abnormal semantic blocks, and dual-view multi-relational edge reconstruction, which detects relation-specific abnormal connections from both intrinsic and behavioral entity profiles. The resulting attribute and relational signals are integrated through a lightweight fusion module to produce the final anomaly score. We further construct 6 benchmark datasets with systematic anomalies, on which extensive experiments show that RelAD consistently outperforms other baselines while achieving competitive efficiency.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The paper proposes RelAD, a reconstruction-based anomaly detection framework for relational databases. It introduces two core modules—conditional sparse-gated attribute reconstruction to suppress redundant multi-table attributes and emphasize abnormal semantic blocks, and dual-view multi-relational edge reconstruction to detect relation-specific abnormal connections from intrinsic and behavioral entity profiles—whose signals are fused to produce anomaly scores. The authors construct 6 benchmark datasets with systematic anomalies and report that RelAD consistently outperforms baselines while maintaining competitive efficiency.

Significance. If the empirical outperformance holds under the reported experimental conditions, the work addresses an under-explored problem in anomaly detection on relational data by jointly modeling attribute heterogeneity and foreign-key connection anomalies. The benchmark construction itself is a useful contribution for the community. The absence of free parameters or invented entities in the method description strengthens the claim that performance derives from the proposed reconstruction objectives rather than hidden fitting.

minor comments (3)
  1. The abstract states outperformance on 6 benchmarks but the experimental section should explicitly tabulate all metrics, baseline descriptions, and hyperparameter settings to allow direct replication; currently the quantitative claims are difficult to verify without those details.
  2. Clarify in §4 how the 6 constructed datasets systematically inject anomalies into foreign-key patterns versus attribute values; the current description leaves open whether the anomaly generation process could inadvertently favor reconstruction-based methods.
  3. The fusion module is described as 'lightweight' but no ablation isolating its contribution versus simple concatenation or averaging appears in the experiments; adding this would strengthen the integration claim.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary, recognition of the problem's importance, and recommendation for minor revision. No specific major comments were raised in the report.

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper introduces RelAD as a reconstruction-based anomaly detection framework consisting of conditional sparse-gated attribute reconstruction and dual-view multi-relational edge reconstruction modules whose outputs are fused into an anomaly score. These components are defined directly from the problem requirements of handling heterogeneous multi-table attributes and foreign-key relations, with no equations or claims reducing a prediction or result back to a fitted parameter or self-citation by construction. Benchmark construction and empirical comparisons are presented as external validation steps rather than inputs that force the method's outputs. No self-citation load-bearing, uniqueness theorem, or ansatz smuggling is indicated in the provided description.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities are specified in the abstract; the report is limited to information extractable from the abstract alone.

pith-pipeline@v0.9.1-grok · 5731 in / 1001 out tokens · 20217 ms · 2026-06-26T21:57:45.625560+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

60 extracted references · 4 canonical work pages · 1 internal anchor

  1. [1]

    Relbench: A benchmark for deep learning on relational databases.Advances in Neural Information Processing Systems, 37:21330–21341, 2024

    Joshua Robinson, Rishabh Ranjan, Weihua Hu, Kexin Huang, Jiaqi Han, Alejandro Dobles, Matthias Fey, Jan E Lenssen, Yiwen Yuan, Zecheng Zhang, et al. Relbench: A benchmark for deep learning on relational databases.Advances in Neural Information Processing Systems, 37:21330–21341, 2024

    2024

  2. [2]

    Relational deep learning: Challenges, foundations and next-generation architectures

    Vijay Prakash Dwivedi, Charilaos Kanatsoulis, Shenyang Huang, and Jure Leskovec. Relational deep learning: Challenges, foundations and next-generation architectures. InProceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V . 2, pages 5999–6009, 2025

    2025

  3. [3]

    Relational graph transformer

    Vijay Prakash Dwivedi, Sri Jaladi, Yangyi Shen, Federico Lopez, Charilaos I Kanatsoulis, Rishi Puri, Matthias Fey, and Jure Leskovec. Relational graph transformer. InTemporal Graph Learning Workshop@ KDD 2025

    2025

  4. [4]

    Relgnn: Composite message passing for relational deep learning

    Tianlang Chen, Charilaos Kanatsoulis, and Jure Leskovec. Relgnn: Composite message passing for relational deep learning. InInternational Conference on Machine Learning, pages 8296–8312. PMLR, 2025

    2025

  5. [5]

    Relational transformer: Toward zero-shot foundation models for relational data

    Rishabh Ranjan, Valter Hudovernik, Mark Znidar, Charilaos I Kanatsoulis, Roshan Reddy Upendra, Mahmoud Mohammadi, Joe Meyer, Tom Palczewski, Carlos Guestrin, and Jure Leskovec. Relational transformer: Toward zero-shot foundation models for relational data. InEurIPS 2025 Workshop: AI for Tabular Data

    2025

  6. [6]

    Griffin: Towards a graph-centric relational database foundation model

    Yanbo Wang, Xiyuan Wang, Quan Gan, Minjie Wang, Qibin Yang, David Wipf, and Muhan Zhang. Griffin: Towards a graph-centric relational database foundation model. InInternational Conference on Machine Learning, pages 64604–64627. PMLR, 2025

    2025

  7. [7]

    Beyond individual input for deep anomaly detection on tabular data

    Hugo Thimonier, Fabrice Popineau, Arpad Rimmel, and Bich-Liên Doan. Beyond individual input for deep anomaly detection on tabular data. InInternational Conference on Machine Learning, pages 48097–48123. PMLR, 2024

    2024

  8. [8]

    Drl: Decomposed representation learning for tabular anomaly detection

    Hangting Ye, He Zhao, Wei Fan, Mingyuan Zhou, Dan dan Guo, and Yi Chang. Drl: Decomposed representation learning for tabular anomaly detection. InThe Thirteenth International Conference on Learning Representations, 2025

    2025

  9. [9]

    Correcting false alarms from unseen: Adapting graph anomaly detectors at test time

    Junjun Pan, Yixin Liu, Chuan Zhou, Fei Xiong, Alan Wee-Chung Liew, and Shirui Pan. Correcting false alarms from unseen: Adapting graph anomaly detectors at test time. InProceedings of the AAAI Conference on Artificial Intelligence, 2026

    2026

  10. [10]

    A survey of generalization of graph anomaly detection: From transfer learning to foundation models

    Junjun Pan, Yu Zheng, Yue Tan, and Yixin Liu. A survey of generalization of graph anomaly detection: From transfer learning to foundation models. InThe 16th IEEE International Conference on Knowledge Graphs, 2025. 10

    2025

  11. [11]

    Deep anomaly detection on attributed networks

    Kaize Ding, Jundong Li, Rohit Bhanushali, and Huan Liu. Deep anomaly detection on attributed networks. InProceedings of the 2019 SIAM international conference on data mining, pages 594–602. SIAM, 2019

    2019

  12. [12]

    Anomaly detection on attributed networks via contrastive self-supervised learning.IEEE transactions on neural networks and learning systems, 33(6):2378–2392, 2021

    Yixin Liu, Zhao Li, Shirui Pan, Chen Gong, Chuan Zhou, and George Karypis. Anomaly detection on attributed networks via contrastive self-supervised learning.IEEE transactions on neural networks and learning systems, 33(6):2378–2392, 2021

    2021

  13. [13]

    Boosting graph anomaly detection with adaptive message passing

    Jingyan Chen, Guanghui Zhu, Chunfeng Yuan, and Yihua Huang. Boosting graph anomaly detection with adaptive message passing. InThe Twelfth International Conference on Learning Representations, 2024

    2024

  14. [14]

    Camera: Adapting to semantic camouflage in unsupervised text-attributed graph fraud detection

    Junjun Pan, Yixin Liu, Yu Zheng, Lianhua Chi, Alan Wee-Chung Liew, and Shirui Pan. Camera: Adapting to semantic camouflage in unsupervised text-attributed graph fraud detection. InInternational Joint Conference on Artificial Intelligence, 2026

    2026

  15. [15]

    Lof: identifying density-based local outliers

    Markus M Breunig, Hans-Peter Kriegel, Raymond T Ng, and Jörg Sander. Lof: identifying density-based local outliers. InProceedings of the 2000 ACM SIGMOD international conference on Management of data, pages 93–104, 2000

    2000

  16. [16]

    Lunar: Unifying local outlier detection methods via graph neural networks

    Adam Goodge, Bryan Hooi, See-Kiong Ng, and Wee Siong Ng. Lunar: Unifying local outlier detection methods via graph neural networks. InProceedings of the AAAI conference on artificial intelligence, volume 36, pages 6737–6745, 2022

    2022

  17. [17]

    Mcm: Masked cell modeling for anomaly detection in tabular data

    Jiaxin Yin, Yuanyuan Qiao, Zitang Zhou, Xiangchao Wang, and Jie Yang. Mcm: Masked cell modeling for anomaly detection in tabular data. InThe Twelfth International Conference on Learning Representations, 2024

    2024

  18. [18]

    Uncertainty- aware graph neural networks: A multihop evidence fusion approach.IEEE Transactions on Neural Networks and Learning Systems, 2025

    Qingfeng Chen, Shiyuan Li, Yixin Liu, Shirui Pan, Geoffrey I Webb, and Shichao Zhang. Uncertainty- aware graph neural networks: A multihop evidence fusion approach.IEEE Transactions on Neural Networks and Learning Systems, 2025

    2025

  19. [19]

    Influence-oriented personalized federated learning.arXiv preprint arXiv:2410.03315, 2024

    Yue Tan, Guodong Long, Jing Jiang, and Chengqi Zhang. Influence-oriented personalized federated learning.arXiv preprint arXiv:2410.03315, 2024

    work page arXiv 2024

  20. [20]

    Arc: A generalist graph anomaly detector with in-context learning.Advances in Neural Information Processing Systems, 37:50772–50804, 2024

    Yixin Liu, Shiyuan Li, Yu Zheng, Qingfeng Chen, Chengqi Zhang, and Shirui Pan. Arc: A generalist graph anomaly detector with in-context learning.Advances in Neural Information Processing Systems, 37:50772–50804, 2024

    2024

  21. [21]

    Freegad: A training- free yet effective approach for graph anomaly detection

    Yunfeng Zhao, Yixin Liu, Shiyuan Li, Qingfeng Chen, Yu Zheng, and Shirui Pan. Freegad: A training- free yet effective approach for graph anomaly detection. InProceedings of the 34th ACM International Conference on Information and Knowledge Management, pages 4379–4389, 2025

    2025

  22. [22]

    RelBench v2: A Large-Scale Benchmark and Repository for Relational Data

    Justin Gu, Rishabh Ranjan, Charilaos Kanatsoulis, Haiming Tang, Martin Jurkovic, Valter Hudovernik, Mark Znidar, Pranshu Chaturvedi, Parth Shroff, Fengyu Li, et al. Relbench v2: A large-scale benchmark and repository for relational data.arXiv preprint arXiv:2602.12606, 2026

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  23. [23]

    Salt: Sales autocompletion linked business tables dataset

    Tassilo Klein, Clemens Biehl, Margarida Costa, Andre Sres, Jonas Kolk, and Johannes Hoffart. Salt: Sales autocompletion linked business tables dataset. InNeurIPS 2024 Third Table Representation Learning Workshop

    2024

  24. [24]

    Prem: A simple yet effective approach for node- level graph anomaly detection

    Junjun Pan, Yixin Liu, Yizhen Zheng, and Shirui Pan. Prem: A simple yet effective approach for node- level graph anomaly detection. In2023 IEEE International Conference on Data Mining (ICDM), pages 1253–1258. IEEE, 2023

    2023

  25. [25]

    Efficient algorithms for mining outliers from large data sets

    Sridhar Ramaswamy, Rajeev Rastogi, and Kyuseok Shim. Efficient algorithms for mining outliers from large data sets. InProceedings of the 2000 ACM SIGMOD international conference on Management of data, pages 427–438, 2000

    2000

  26. [26]

    Generative adversarial active learning for unsupervised outlier detection.IEEE Transactions on Knowledge and Data Engineering, 32(8):1517–1528, 2019

    Yezheng Liu, Zhe Li, Chong Zhou, Yuanchun Jiang, Jianshan Sun, Meng Wang, and Xiangnan He. Generative adversarial active learning for unsupervised outlier detection.IEEE Transactions on Knowledge and Data Engineering, 32(8):1517–1528, 2019

    2019

  27. [27]

    Relational deep learning: Graph representation learning on relational databases

    Joshua Robinson, Rishabh Ranjan, Weihua Hu, Kexin Huang, Jiaqi Han, Alejandro Dobles, Matthias Fey, Jan Eric Lenssen, Yiwen Yuan, Zecheng Zhang, et al. Relational deep learning: Graph representation learning on relational databases. InNeurIPS 2024 third table representation learning workshop, 2024

    2024

  28. [28]

    Large language models are good relational learners

    Fang Wu, Vijay Prakash Dwivedi, and Jure Leskovec. Large language models are good relational learners. InProceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7835–7854, 2025. 11

    2025

  29. [29]

    Plurel: Synthetic data unlocks scaling laws for relational foundation models

    Vignesh Kothapalli, Rishabh Ranjan, Valter Hudovernik, Vijay Prakash Dwivedi, Johannes Hoffart, Carlos Guestrin, and Jure Leskovec. Plurel: Synthetic data unlocks scaling laws for relational foundation models. arXiv preprint arXiv:2602.04029, 2026

    work page arXiv 2026

  30. [30]

    Deep learning for anomaly detection: A review.ACM computing surveys (CSUR), 54(2):1–38, 2021

    Guansong Pang, Chunhua Shen, Longbing Cao, and Anton Van Den Hengel. Deep learning for anomaly detection: A review.ACM computing surveys (CSUR), 54(2):1–38, 2021

    2021

  31. [31]

    Deep neural networks and tabular data: A survey.IEEE transactions on neural networks and learning systems, 35(6):7499–7519, 2022

    Vadim Borisov, Tobias Leemann, Kathrin Seßler, Johannes Haug, Martin Pawelczyk, and Gjergji Kasneci. Deep neural networks and tabular data: A survey.IEEE transactions on neural networks and learning systems, 35(6):7499–7519, 2022

    2022

  32. [32]

    Anomaly detection for tabular data with internal contrastive learning

    Tom Shenkar and Lior Wolf. Anomaly detection for tabular data with internal contrastive learning. In International conference on learning representations, 2022

    2022

  33. [33]

    A comprehensive survey on graph anomaly detection with deep learning.IEEE transactions on knowledge and data engineering, 35(12):12012–12038, 2021

    Xiaoxiao Ma, Jia Wu, Shan Xue, Jian Yang, Chuan Zhou, Quan Z Sheng, Hui Xiong, and Leman Akoglu. A comprehensive survey on graph anomaly detection with deep learning.IEEE transactions on knowledge and data engineering, 35(12):12012–12038, 2021

    2021

  34. [34]

    Deep graph anomaly detection: A survey and new perspectives.IEEE Transactions on Knowledge and Data Engineer- ing, 2025

    Hezhe Qiao, Hanghang Tong, Bo An, Irwin King, Charu Aggarwal, and Guansong Pang. Deep graph anomaly detection: A survey and new perspectives.IEEE Transactions on Knowledge and Data Engineer- ing, 2025

    2025

  35. [35]

    From few-shot to zero-shot: Towards generalist graph anomaly detection.IEEE Transactions on Knowledge and Data Engineering, 2026

    Yixin Liu, Shiyuan Li, Yu Zheng, Qingfeng Chen, Chengqi Zhang, Philip S Yu, and Shirui Pan. From few-shot to zero-shot: Towards generalist graph anomaly detection.IEEE Transactions on Knowledge and Data Engineering, 2026

    2026

  36. [36]

    Raising the bar in graph ood generalization: Invariant learning beyond explicit environment modeling.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026

    Xu Shen, Yixin Liu, Yili Wang, Rui Miao, Yiwei Dai, Shirui Pan, Yi Chang, and Xin Wang. Raising the bar in graph ood generalization: Invariant learning beyond explicit environment modeling.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026

    2026

  37. [37]

    Fedcigar: A personalized reconstruction approach for federated graph-level anomaly detection

    Yunfeng Zhao, Yixin Liu, Qingfeng Chen, Shiyuan Li, Yue Tan, and Shirui Pan. Fedcigar: A personalized reconstruction approach for federated graph-level anomaly detection. InInternational Joint Conference on Artificial Intelligence, 2026

    2026

  38. [38]

    Rethinking feature alignment in generalist graph anomaly detection: A relational fingerprint-based approach

    Yujing Liu, Yixin Liu, Yu Zheng, Alan Wee-Chung Liew, Xiaofeng Cao, and Shirui Pan. Rethinking feature alignment in generalist graph anomaly detection: A relational fingerprint-based approach. In International Conference on Machine Learning, 2026

    2026

  39. [39]

    Anomalous: A joint modeling approach for anomaly detection on attributed networks

    Zhen Peng, Minnan Luo, Jundong Li, Huan Liu, Qinghua Zheng, et al. Anomalous: A joint modeling approach for anomaly detection on attributed networks. InIjcai, volume 18, pages 3513–3519, 2018

    2018

  40. [40]

    Radar: residual analysis for anomaly detection in attributed networks

    Jundong Li, Harsh Dani, Xia Hu, and Huan Liu. Radar: residual analysis for anomaly detection in attributed networks. InProceedings of the 26th International Joint Conference on Artificial Intelligence, pages 2152–2158, 2017

    2017

  41. [41]

    Truncated affinity maximization: One-class homophily modeling for graph anomaly detection.Advances in Neural Information Processing Systems, 36:49490–49512, 2023

    Hezhe Qiao and Guansong Pang. Truncated affinity maximization: One-class homophily modeling for graph anomaly detection.Advances in Neural Information Processing Systems, 36:49490–49512, 2023

    2023

  42. [42]

    Modeling relational data with graph convolutional networks

    Michael Schlichtkrull, Thomas N Kipf, Peter Bloem, Rianne Van Den Berg, Ivan Titov, and Max Welling. Modeling relational data with graph convolutional networks. InEuropean semantic web conference, pages 593–607. Springer, 2018

    2018

  43. [43]

    Blindguard: Safeguarding llm-based multi-agent systems under unknown attacks

    Rui Miao, Yixin Liu, Yili Wang, Xu Shen, Yue Tan, Yiwei Dai, Shirui Pan, and Xin Wang. Blindguard: Safeguarding llm-based multi-agent systems under unknown attacks. InProceedings of the 64th Annual Meeting of the Association for Computational Linguistics, 2026

    2026

  44. [44]

    Dynhd: Hallucination detection for diffusion large language models via denoising dynamics deviation learning.arXiv preprint arXiv:2603.16459, 2026

    Yanyu Qian, Yue Tan, Yixin Liu, Wang Yu, and Shirui Pan. Dynhd: Hallucination detection for diffusion large language models via denoising dynamics deviation learning.arXiv preprint arXiv:2603.16459, 2026

    work page arXiv 2026

  45. [45]

    Assemble your crew: Automatic multi-agent communication topology design via autoregressive graph generation

    Shiyuan Li, Yixin Liu, Qingsong Wen, Chengqi Zhang, and Shirui Pan. Assemble your crew: Automatic multi-agent communication topology design via autoregressive graph generation. InProceedings of the AAAI Conference on Artificial Intelligence, volume 40, pages 23142–23150, 2026

    2026

  46. [46]

    Ofa-mas: One-for-all multi-agent system topology design based on mixture-of-experts graph generative models

    Shiyuan Li, Yixin Liu, Yu Zheng, Mei Li, Quoc Viet Hung Nguyen, and Shirui Pan. Ofa-mas: One-for-all multi-agent system topology design based on mixture-of-experts graph generative models. InProceedings of the ACM Web Conference 2026, pages 1333–1344, 2026

    2026

  47. [47]

    Rethinking unsupervised time series anomaly detection: Dynamic attention based on route inverse-masking.Applied Soft Computing, page 113971, 2025

    Enguang Zuo, Jie Zhong, Chen Chen, Cheng Chen, Kurban Ubul, and Xiaoyi Lv. Rethinking unsupervised time series anomaly detection: Dynamic attention based on route inverse-masking.Applied Soft Computing, page 113971, 2025. 12

    2025

  48. [48]

    Enguang Zuo, Junyi Yan, Alimjan Aysa, Chen Chen, Cheng Chen, Hongbing Ma, Xiaoyi Lv, and Kurban Ubul. Sucola: Self-adaptive structure refinement unsupervised contrastive learning framework for food safety risk early warning.Engineering Applications of Artificial Intelligence, 126:107016, 2023

    2023

  49. [49]

    Retrieval augmented deep anomaly detection for tabular data

    Hugo Thimonier, Fabrice Popineau, Arpad Rimmel, and Bich-Liên Doan. Retrieval augmented deep anomaly detection for tabular data. InProceedings of the 33rd ACM international conference on information and knowledge management, pages 2250–2259, 2024

    2024

  50. [50]

    Anollm: Large language models for tabular anomaly detection

    Che-Ping Tsai, Ganyu Teng, Phillip Wallis, and Wei Ding. Anollm: Large language models for tabular anomaly detection. InThe Thirteenth International Conference on Learning Representations, 2025

    2025

  51. [51]

    Disentangling tabular data towards better one-class anomaly detection

    Jianan Ye, Zhaorui Tan, Yijie Hu, Xi Yang, Guangliang Cheng, and Kaizhu Huang. Disentangling tabular data towards better one-class anomaly detection. InProceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 13061–13068, 2025

    2025

  52. [52]

    Data-efficient and interpretable tabular anomaly detection

    Chun-Hao Chang, Jinsung Yoon, Sercan Ö Arik, Madeleine Udell, and Tomas Pfister. Data-efficient and interpretable tabular anomaly detection. InProceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 190–201, 2023

    2023

  53. [53]

    Towards one-for-all anomaly detection for tabular data

    Shiyuan Li, Yixin Liu, Yu Zheng, Xiaofeng Cao, Shirui Pan, and Heng Tao Shen. Towards one-for-all anomaly detection for tabular data. InForty-third international conference on machine learning, 2026

    2026

  54. [54]

    Graph anomaly detection in time series: A survey.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025

    Thi Kieu Khanh Ho, Ali Karami, and Narges Armanfard. Graph anomaly detection in time series: A survey.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025

    2025

  55. [55]

    Address anomalies at critical crossroads for graph anomaly detection.IEEE Transactions on Knowledge and Data Engineering, 2025

    Junyi Yan, Enguang Zuo, Ke Liang, Meng Liu, Miaomiao Li, Xinwang Liu, Xiaoyi Lv, and Kai Lu. Address anomalies at critical crossroads for graph anomaly detection.IEEE Transactions on Knowledge and Data Engineering, 2025

    2025

  56. [56]

    Smoothgnn: Smoothing-aware gnn for unsupervised node anomaly detection

    Xiangyu Dong, Xingyi Zhang, Yanni Sun, Lei Chen, Mingxuan Yuan, and Sibo Wang. Smoothgnn: Smoothing-aware gnn for unsupervised node anomaly detection. InProceedings of the ACM on Web Conference 2025, pages 1225–1236, 2025

    2025

  57. [57]

    verified purchase

    Rui Bing, Guan Yuan, Mu Zhu, Fanrong Meng, Huifang Ma, and Shaojie Qiao. Heterogeneous graph neural networks analysis: a survey of techniques, evaluations and applications.Artificial Intelligence Review, 56(8), 2023. 13 A Algorithm Description The training and testing algorithms are given in Algorithm 1 and Algorithm 2, respectively. Algorithm 1:The Train...

    2023

  58. [58]

    Importantly, anomaly labels are assigned only after verifying that the sampled entities satisfy the dataset-specific injection constraints

    Stratified Injection Rate:We initially sample 5% of the total population across all datasets using a stratified sampling strategy (e.g., binning by node activity or degree) to eliminate selection bias, ensuring that the statistical distribution of the anomalous group rigorously aligns with that of the normal group. Importantly, anomaly labels are assigned...

  59. [59]

    Replace-Only

    Strict “Replace-Only” Strategy:We strictly enforce a “replace-only” strategy during the injection phase, firmly prohibiting the addition or deletion of any data rows. This guarantees that the foundational statistical features of the nodes (e.g., interaction frequency, total degree) remain absolutely conserved before and after injection

  60. [60]

    Guided by these principles, we design a separate injection rule for each dataset according to its specific schema and business context

    Real-World Scenario Reconstruction:Each injection strategy is explicitly designed to reconstruct real-world business fraud scenarios, discarding mere random noise. Guided by these principles, we design a separate injection rule for each dataset according to its specific schema and business context. The per-dataset details are as follows: • Amazon.In e-com...