Pith. sign in

REVIEW 2 major objections 6 minor 57 references

GuARD: Effective Anomaly Detection through a Text-Rich and Graph-Informed Language Model

T0 review · 2 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Feeding an LLM a text token and a graph token makes it detect anomalies better and up to 10x faster.

desk verdict Neat integration of SLM semantic and GNN structural tokens into a LoRA-tuned LLM, but the multi-turn training/inference mismatch in the placeholder labels needs to be resolved before the multi-turn accuracy gains are taken at face value. read the letter →

arxiv 2412.03930 v2 pith:GAPBG2X5 submitted 2024-12-05 cs.CL cs.AI

Yunhe Pang , Bo Chen , Fanjin Zhang , Yanghui Rao , Evgeny Kharlamov , Jie Tang This is my paper · ORCID
classification cs.CLcs.AI
keywords anomalydetectiontext-richgraphslargelanguagemodelsinstructiontuninggraphneuralnetworksauthornamedisambiguationmulti-turnsemanticembedding
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

GuARD is a method for anomaly detection on text-rich graphs—networks where each node carries long text, such as an author's papers or a social media account. The paper claims that a large language model fine-tuned with a multi-turn instruction template, then augmented with two special input tokens—one summarizing each node's full text via a small language model, and one encoding the node's graph neighborhood via a GNN—beats both graph-based and LLM-based baselines on four datasets. It further claims that this design avoids the cost of long-context LLM fine-tuning, delivering up to 5x faster training and 3x–10x faster inference than LoRA-tuned full-text LLMs. If true, it makes LLM-powered anomaly detection practical on large academic and social graphs.

What carries the argument

The load-bearing mechanism is the multi-turn instruction template together with the two special tokens. The template stacks several target nodes into one prompt, sharing a global context, so the causal LLM decodes a <label_token> for each turn in a single autoregressive pass; training minimizes the sum of per-turn label log-likelihoods, and inference reads the normalized logits of 'Yes' versus 'No'. The <text> token inserts a mean-pooled small-language-model embedding of the node's full attributes, and the <graph> token inserts a GNN-derived embedding concatenated with a graph-level embedding, each aligned to the LLM hidden space by a two-layer Swish FFN. Three progressive stages freeze earlier parameters in turn, which the paper argues yields more coherent fusion than jointly training both projectors.

What would settle it

Run GuARD with the multi-turn template at inference but replace the ground-truth label tokens of earlier turns with the model's own predicted tokens; if the 8-turn AUC on WhoIsWho falls to the 1-turn level (0.744), the multi-turn improvement is an artifact of label leakage rather than context sharing.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that the three ingredients—task-guided multi-turn instruction tuning, a semantic embedding module, and a structural embedding module—combine progressively to fuse key text, rich text, and graph topology. The base model learns to answer 'is this node an outlier?' by reading stacked queries with a shared global context; later stages replace the special <text> token with a mean-pooled SLM embedding and the <graph> token with a GNN node-plus-graph embedding, while freezing earlier parameters. The resulting GuARD+graph variant reaches state-of-the-art AUC on WhoIsWho (0.789), MAG (0.963), TwiBot-20 (0.945), and SemEval-23F (0.875), and in the WhoIsWho KDD Cup 2024 test leaderboard comparison it achieves 83.51% AUC in 1.3 hours of test time, where the top LLM-based solutions needed over 10 hours.

Load-bearing premise

The multi-turn accuracy gain assumes that training with ground-truth labels present in earlier turns of the context transfers to inference, where those labels are absent and the model must rely on its own predictions.

Editorial extensions

If this is right

  • LLM-based anomaly detection can scale to graphs with thousands of long-text nodes without truncating away detection cues.
  • Structural information can be injected into a frozen, instruction-tuned LLM through a single token, without retraining the backbone.
  • Multi-turn stacking turns one decoding pass into many node predictions, cutting training and inference time proportionally to the number of turns.
  • The same recipe transfers across domains: author-name disambiguation, bot detection, and misinformation/propaganda identification.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the multi-turn gain is genuine, the same teacher-forcing concern applies: training sees ground-truth labels in earlier turns, but inference does not, so the 8-turn improvement over 1-turn (0.763 vs 0.744 AUC on WhoIsWho) may partly reflect label leakage rather than true context sharing. Filling earlier label positions with the model's own predictions at inference would test this.
  • The graph token framework is agnostic to the GNN used, so any structural encoder (e.g., heterogeneous transformers or peer-aware modules) could plug in; GuARD's gains on TwiBot-20 over SLM features suggest text and structure are complementary rather than redundant.
  • The speedups imply that per-node anomaly scores could be computed in near-real time for streaming social graphs, a deployment regime the paper does not discuss.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 6 minor

Summary. The paper proposes GuARD, an anomaly-detection method for text-rich graphs. It combines an instruction-tuned LLM (Llama3-8B or Qwen2.5-7B) with two auxiliary modalities: a mean-pooled small-language-model text embedding injected as a special <text> token, and a GNN (GCCAD/HGT) structural embedding injected as a <graph> token. The model is trained in three progressive stages: first, a base LLM is fine-tuned with a multi-turn instruction template in which several target nodes are scored in one context; second, a text projector is trained while the LLM is frozen; third, a graph projector is trained while all earlier parameters are frozen. Experiments on WhoIsWho, MAG, TwiBot-20, and SemEval-23F report AUC/MAP gains over GNN, SLM, and long-context LLM baselines, together with large training and inference speedups over LoRA-fine-tuned Llama3-8B. The paper also includes extensive ablations over the number of turns, text projectors, training order, foundation models, and paper attributes.

Significance. If the reported results hold, GuARD is a practically valuable contribution: it shows that a relatively small set of key text tokens plus compressed semantic and structural tokens can match or exceed full-context LLM fine-tuning at a fraction of the cost. The three-stage progressive training recipe and the use of special tokens to inject non-linguistic modalities are clean and reproducible ideas. The strengths of the paper include its public code release, the breadth of the evaluation across four datasets, and the systematic ablations (Tables 2-5, Figures 4-8). The main caveat is the train/inference protocol for the multi-turn template, which must be specified and validated before the central empirical claim is secure.

major comments (2)
  1. [§4.1, Eq. (1)] The training objective is a sum over N label positions, and the text states that earlier queries 'can serve as few-shot examples for the prediction of label tokens in latter queries.' This implies that during training the context for turn i contains the ground-truth label tokens of turns 1..i-1 (teacher forcing). The inference description in Eqs. (1)-(2) does not say whether earlier <label_token> positions are filled with model-predicted labels before later positions are scored. If they are left as the literal placeholder token, the inference-time context distribution (only placeholder embeddings) differs from the training-time distribution (ground-truth labels), so the multi-turn accuracy gain in Table 4 (WhoIsWho AUC 0.744 at 1 turn vs. 0.763 at 8 turns) and the speedups in Table 5 may be artifacts of teacher-forced label leakage. Please specify the exact inference protocol (parallel placeholder scoring vs. sequential insertion of decoded labels), match it to the training objective, and re-report Table 4 and Table 5 under that protocol.
  2. [§5.5, Table 5] The claimed speedup numbers are internally inconsistent. Table 5 reports WhoIsWho training time 39.00 min for GuARD+graph vs. 592.80 min for Llama3-8B, which is a 15.2x training speedup, while the abstract says 'up to 5x speedup in training' and Section 5.5 says 'over a 10x speedup in inference and a 5x speedup in training.' The paper should state the correct maximum speedups and define the comparison protocol (LoRA fine-tuning vs. full fine-tuning) consistently in the abstract, Section 5.5, and Table 5.
minor comments (6)
  1. [§4.1, Eq. (1)] The notation 'w_i denotes the logits of the ground-truth label' is imprecise: in log p(w_i | context_i), w_i should be the ground-truth label token, while the model output logits are z_i (used in Eq. (2)). Please clarify the notation.
  2. [Abstract] The phrase 'pave a new revenue' should be 'pave a new avenue.'
  3. [§5.2, Table 4] The number of turns is set to 10 for WhoIsWho and MAG, but Table 4 reports the best WhoIsWho AUC at 8 turns (0.763) and a degradation at 16 turns (0.737); the choice of 10 is not justified by the ablation. Either report the 10-turn result or align the configuration with the ablation.
  4. [§5.4] There are several typos: 'reserve of two-modal training' should be 'reverse of two-modal training', 'As as result' in §4.2 should be 'As a result', and 'intergrate' in §4.4 should be 'integrate.'
  5. [§5.3, Table 6] Table 6 reports GuARD AUC 83.51%, which differs from the 0.789 AUC in Table 2; the text in Section 5.6 should explicitly state that Table 6 uses the ensemble (T+TO+TA) on the KDD Cup test set, not the single GuARD+graph model, to avoid confusion.
  6. [§4.3, §5.1] The paper does not state whether the GNN structural encoder (GCCAD/HGT) is trained on the training split only or on the full graph including test nodes; since graph learning is often transductive, this should be clarified to rule out label leakage through the structural embeddings.

Circularity Check

0 steps flagged · score 2.0 of 10

No material circularity: GuARD's gains are empirical comparisons against external baselines; self-citations to WhoIsWho and GCCAD are component/benchmark reuse, not forced reductions.

full rationale

GuARD's central claims are empirical rather than derivational. The model is assembled from published components (a frozen small language model for semantic tokens, a separately trained GNN encoder for graph tokens, and an instruction-tuned LLM) and evaluated on held-out test sets of four datasets, including the external MAG, TwiBot-20, and SemEval-23F benchmarks. The authors' own WhoIsWho benchmark and GCCAD encoder are reused, but the paper does not define GuARD's output as those components' outputs: the semantic and graph tokens are inputs to a further instruction-tuned LLM, and the reported improvements over GCCAD, RoBERTa, DeBERTa, Llama3-8B, and Qwen2.5-7B baselines are measured outcomes, not identities. The multi-turn template's teacher-forced training objective in Eq. (1) versus the placeholder-based logit reading in Eq. (2) is a genuine train/inference mismatch and a correctness risk, but it is not a circular reduction: no equation is equal to an input by construction, and no fitted parameter is renamed as a prediction. The self-citations are minor and non-load-bearing, so the circularity score is low.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on standard ML engineering choices (LoRA, PLM pooling, GNN embeddings) plus several domain assumptions about the multi-turn template and the information content of pooled embeddings. No exotic entities are introduced; the only new artifacts are architectural tokens, which are part of the method rather than explanatory postulates.

free parameters (3)
  • per-dataset number of instruction turns = WhoIsWho: 10, MAG: 10, TwiBot-20: 8, SemEval-23F: 6
    Chosen via ablation on WhoIsWho and TwiBot-20 (Table 4) and set for other datasets; directly affects accuracy and the speedup numbers.
  • key textual attributes per dataset = WhoIsWho: Title+Author; MAG: Title+Author (Appendix B); TwiBot-20: metadata; SemEval-23F: first 512 tokens
    Selected using ablations on the same datasets (Figures 4 and 9); the final reported results use these hand-picked inputs.
  • LoRA rank and alpha = rank=8, alpha=16, dropout=0.05
    Standard configuration, consistent across experiments; not fitted to the anomaly detection target.
assumptions (4)
  • domain assumption Causal attention over stacked queries lets earlier queries serve as useful few-shot demonstrations for later predictions.
    Section 4.1 asserts this mechanism; training (Eq. 1) teacher-forces ground-truth labels for earlier turns, which are not available at inference, so the assumption is not verified under deployment conditions.
  • domain assumption Mean pooling of all PLM token embeddings preserves the semantic information needed for anomaly detection.
    Section 4.2 (Eqs. 3-5) uses mean pooling over the text encoder outputs; no analysis shows that rare but discriminative cues survive averaging.
  • domain assumption A frozen GNN encoder trained on the same graph provides structural features complementary to the text tokens.
    Section 4.3-4.4 freezes the GNN and trains only a projector; complementarity is inferred from final accuracy, not directly tested.
  • domain assumption Randomly sampled global context nodes give a representative reference set for judging a target node.
    Section 4.1 randomly samples a fixed number of global context nodes; the paper does not analyze sensitivity to the sampling.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GuARD: Effective Anomaly Detection through a Text-Rich and Graph-Informed Language Model." pith.science (2026). https://pith.science/paper/GAPBG2X5

@misc{pith2026241203930,
  author       = {Pith},
  title        = {Pith review of: GuARD: Effective Anomaly Detection through a Text-Rich and Graph-Informed Language Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GAPBG2X5}},
  note         = {Machine review of arXiv:2412.03930}
}
abstract

Anomaly detection on text-rich graphs is widely prevalent in real life, such as detecting incorrectly assigned academic papers to authors and detecting bots in social networks. The remarkable capabilities of large language models (LLMs) pave a new revenue by utilizing rich-text information for effective anomaly detection. However, simply introducing rich texts into LLMs can obscure essential detection cues and introduce high fine-tuning costs. Moreover, LLMs often overlook the intrinsic structural bias of graphs which is vital for distinguishing normal from abnormal node patterns. To this end, this paper introduces GuARD, a text-rich and graph-informed language model that combines key structural features from graph-based methods with fine-grained semantic attributes extracted via small language models for effective anomaly detection on text-rich graphs. GuARD is optimized with the progressive multi-modal multi-turn instruction tuning framework in the task-guided instruction tuning regime tailed to incorporate both rich-text and structural modalities. Extensive experiments on four datasets reveal that GuARD outperforms graph-based and LLM-based anomaly detection methods, while offering up to 5$\times$ times speedup in training and 5$\times$ times speedup in inference over vanilla long-context LLMs on the large-scale WhoIsWho dataset.

Figures

Figures reproduced from arXiv: 2412.03930 by the authors.

Figure 2
Figure 2. AUC vs. inference time of different models on the [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 1
Figure 1. Illustration about typical anomaly detection in text [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. The entire workflow of GuARD in detecting anomalies on text-rich graphs. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Ablation studies on different paper attributes as the [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Ablation studies on the different text projectors [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: The AUC comparison of different training strategies [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Comparison between individually trained models [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Performance comparison of different ensemble [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]
Figure 9
Figure 9. Figure 9: Ablation studies on different paper attributes as the [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 10
Figure 10. Figure 10: Comparison of different model scales (3B, 7B, 14B) [PITH_FULL_IMAGE:figures/full_fig_p012_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 35 canonical work pages

  1. [1]

    Sepideh Bazzaz Abkenar, Mostafa Haghi Kashani, Mohammad Akbari, and Ebrahim Mahdipour. 2023. Learning textual features for Twitter spam detection: A systematic literature review. Expert Systems with Applications 228 (2023)

    work page 2023

  2. [2]

    Yushi Bai, Xin Lv, Jiajie Zhang, Yuze He, Ji Qi, Lei Hou, Jie Tang, Yuxiao Dong, and Juanzi Li. 2024. LongAlign: A recipe for long context alignment of large KDD ’25, August 3–7, 2025, Toronto, ON, Canada Trovato et al. language models. In Findings of the 2024 Conference on Empirical Methods in Natural Language Processing

    work page 2024

  3. [3]

    Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. 2020. Language models are few-shot learners. Advances in Neural Information Processing Systems 33 (2020), 1877–1901

    2020

  4. [4]

    Zijian Cai, Zhaoxuan Tan, Zhenyu Lei, Zifeng Zhu, Hongrui Wang, Qinghua Zheng, and Minnan Luo. 2024. LMBot: Distilling Graph Knowledge into Language Model for Graph-less Deployment in Twitter Bot Detection. In Proceedings of the 17th ACM International Conference on Web Search and Data Mining . 57–66

    work page 2024

  5. [5]

    Bo Chen, Jing Zhang, Fanjin Zhang, Tianyi Han, Yuqing Cheng, Xiaoyan Li, Yuxiao Dong, and Jie Tang. 2023. Web-scale academic name disambiguation: the WhoIsWho benchmark, leaderboard, and toolkit. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . 3817–3828

    work page 2023

  6. [6]

    Bo Chen, Jing Zhang, Xiaokang Zhang, Yuxiao Dong, Jian Song, Peng Zhang, Kaibo Xu, Evgeny Kharlamov, and Jie Tang. 2022. GCCAD: Graph contrastive cod- ing for anomaly detection. IEEE Transactions on Knowledge and Data Engineering 35 (2022), 8037–8051

    work page 2022

  7. [7]

    Canyu Chen and Kai Shu. 2024. Can LLM-Generated misinformation be detected?. In Proceeding of the 12th International Conference on Learning Representations

    work page 2024

  8. [8]

    Tri Dao. 2023. Flashattention-2: Faster attention with better parallelism and work partitioning. arXiv preprint arXiv:2307.08691 (2023)

    arXiv 2023

Show all 57 references
  1. [9]

    Tri Dao, Dan Fu, Stefano Ermon, Atri Rudra, and Christopher Ré. 2022. Flashat- tention: Fast and memory-efficient exact attention with io-awareness. Advances in Neural Information Processing Systems 35 (2022), 16344–16359

    2022

  2. [10]

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding.arXiv preprint arXiv:1810.04805 (2018)

    2018 arXiv

  3. [11]

    Yingtong Dou, Zhiwei Liu, Li Sun, Yutong Deng, Hao Peng, and Philip S. Yu. 2020. Enhancing Graph Neural Network-based Fraud Detectors against Camouflaged Fraudsters. In Proceeding of the 29th ACM International Conference on Information and Knowledge Management

    2020

  4. [12]

    Shangbin Feng, Herun Wan, Ningnan Wang, Jundong Li, and Minnan Luo. 2021. Twibot-20: A comprehensive twitter bot detection benchmark. In Proceedings of the 30th ACM international conference on information and knowledge management . 4485–4494

    2021

  5. [13]

    Shangbin Feng, Herun Wan, Ningnan Wang, Zhaoxuan Tan, Minnan Luo, and Yulia Tsvetkov. 2024. What does the bot say? Opportunities and risks of large language models in social media bot detection. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics

    2024

  6. [14]

    Team GLM, Aohan Zeng, Bin Xu, Bowen Wang, Chenhui Zhang, Da Yin, Diego Rojas, Guanyu Feng, Hanlin Zhao, Hanyu Lai, et al. 2024. ChatGLM: A family of large language models from GLM-130B to GLM-4 all tools. arXiv preprint arXiv:2406.12793 (2024)

    2024 arXiv

  7. [15]

    Wei Guan, Jian Cao, Shiyou Qian, and Jianqi Gao. 2024. LogLLM: Log-based anomaly detection using large language models. arXiv preprint arXiv:2411.08561 (2024)

    2024 arXiv

  8. [16]

    Zeyu Han, Chao Gao, Jinyang Liu, Sai Qian Zhang, et al . 2024. Parameter- efficient fine-tuning for large models: A comprehensive survey. arXiv preprint arXiv:2403.14608 (2024)

    2024 arXiv

  9. [17]

    Pengcheng He, Xiaodong Liu, Jianfeng Gao, and Weizhu Chen. 2021. Deberta: decoding-Enhanced Bert with Disentangled Attention. In Proceeding of the 9th International Conference on Learning Representations

    2021

  10. [18]

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2021. LoRA: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685 (2021)

    2021 arXiv

  11. [19]

    Ziniu Hu, Yuxiao Dong, Kuansan Wang, and Yizhou Sun. 2020. Heterogeneous Graph Transformer. In Proceedings of the Web Conference 2020 . 2704–2710

    2020

  12. [20]

    Albert Q Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, De- vendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, et al . 2023. Mistral 7B. arXiv preprint arXiv:2310.06825 (2023)

    2023 arXiv

  13. [21]

    Wei Ju, Zhengyang Mao, Siyu Yi, Yifang Qin, Yiyang Gu, Zhiping Xiao, Yifan Wang, Xiao Luo, and Ming Zhang. 2024. Hypergraph-enhanced Dual Semi- supervised Graph Classification. In Proceeding of the 41st International Conference on Machine Learning

    2024

  14. [22]

    Guolin Ke, Qi Meng, Thomas Finley, Taifeng Wang, Wei Chen, Weidong Ma, Qiwei Ye, and Tie-Yan Liu. 2017. Lightgbm: A highly efficient gradient boosting decision tree. Advances in Neural Information Processing Systems 30 (2017)

    2017

  15. [23]

    Thomas N Kipf and Max Welling. 2017. Semi-supervised classification with graph convolutional networks. (2017)

    2017

  16. [24]

    Mario Michael Krell, Matej Kosec, Sergio P Perez, and Andrew Fitzgibbon. 2021. Efficient sequence packing without cross-contamination: Accelerating large lan- guage models without impacting performance. arXiv preprint arXiv:2107.02027 (2021)

    2021 arXiv

  17. [25]

    Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. 2023. Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models. In Proceedings of the 40th International Conference on Machine Learning . 19730–19742

    2023

  18. [26]

    Xiang Lisa Li and Percy Liang. 2021. Prefix-tuning: optimizing continuous prompts for generation. In Proceedings of the 59th Annual Meeting of the Associa- tion for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. 4582–4597

    2021

  19. [27]

    Zehan Li, Xin Zhang, Yanzhao Zhang, Dingkun Long, Pengjun Xie, and Meishan Zhang. 2023. Towards general text embeddings with multi-stage contrastive learning. arXiv preprint arXiv:2308.03281 (2023)

    2023 arXiv

  20. [28]

    Liu, Kevin Lin, John Hewitt, Ashwin Paranjape, Michele Bevilacqua, Fabio Petroni, and Percy Liang

    Nelson F. Liu, Kevin Lin, John Hewitt, Ashwin Paranjape, Michele Bevilacqua, Fabio Petroni, and Percy Liang. 2024. Lost in the middle: How language models use long contexts. Transactions of the Association for Computational Linguistics 12 (2024)

    2024

  21. [29]

    Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692 (2019)

    2019 arXiv

  22. [30]

    Ilya Loshchilov and Frank Hutter. 2017. Fixing weight decay regularization in adam. CoRR abs/1711.05101 (2017)

    2017 arXiv

  23. [31]

    Jason Samuel Lucas, Adaku Uchendu, Michiharu Yamashita, Jooyoung Lee, Shau- rya Rohatgi, and Dongwon Lee. 2023. Fighting fire with fire: The dual role of LLMs in crafting and detecting elusive disinformation. In Proceedings of the 2023 Conference on Empirical Methods in Natura...

    2023

  24. [32]

    Qingsong Lv, Ming Ding, Qiang Liu, Yuxiang Chen, Wenzheng Feng, Siming He, Chang Zhou, Jianguo Jiang, Yuxiao Dong, and Jie Tang. 2021. Are we really making much progress?: Revisiting, benchmarking and refining heterogeneous graph neural networks. In Proceeding of the 27th ACM ...

    2021

  25. [33]

    Sheng, Hui Xiong, and Leman Akoglu

    Xiaoxiao Ma, Jia Wu, Shan Xue, Jian Yang, Chuan Zhou, Quan Z. Sheng, Hui Xiong, and Leman Akoglu. 2023. A Comprehensive Survey on Graph Anom- aly Detection With Deep Learning. IEEE Transactions on Knowledge and Data Engineering 35 (2023)

    2023

  26. [34]

    AI Meta. 2024. Introducing meta llama 3: The most capable openly available llm to date. Meta AI (2024)

    2024

  27. [35]

    Jakub Piskorski, Nicolas Stefanovitch, Giovanni Da San Martino, and Preslav Nakov. 2023. Semeval-2023 task 3: Detecting the category, the framing, and the persuasion techniques in online news in a multi-lingual setup. In Proceedings of the 17th International Workshop on Semant...

    2023

  28. [36]

    Prajit Ramachandran, Barret Zoph, and Quoc V Le. 2017. Searching for activation functions. arXiv preprint arXiv:1710.05941 (2017)

    2017 arXiv

  29. [37]

    Senjuti Basu Roy, Martine De Cock, Vani Mandava, Swapna Savanna, Brian Dalessandro, Claudia Perlich, William Cukierski, and Ben Hamner. 2013. The microsoft academic search dataset and kdd cup 2013. In Proceedings of the 2013 KDD cup 2013 workshop . 1–6

    2013

  30. [38]

    Kipf, Peter Bloem, Rianne van den Berg, Ivan Titov, and Max Welling

    Michael Sejr Schlichtkrull, Thomas N. Kipf, Peter Bloem, Rianne van den Berg, Ivan Titov, and Max Welling. 2018. Modeling Relational Data with Graph Convo- lutional Networks. In The semantic web: 15th international conference , Vol. 10843

    2018

  31. [39]

    Noam Shazeer. 2020. Glu variants improve transformer. arXiv preprint arXiv:2002.05202 (2020)

    2020 arXiv

  32. [40]

    Ming Shen. 2024. An ensemble model with multi-scale features for incorrect assignment detection. In KDD 2024 OAG-Challenge Cup

    2024

  33. [41]

    Jianheng Tang, Fengrui Hua, Ziqi Gao, Peilin Zhao, and Jia Li. 2023. GADBench: Revisiting and Benchmarking Supervised Graph Anomaly Detection. InAdvances in Neural Information Processing Systems , Vol. 36. 29628–29653

    2023

  34. [42]

    Jianheng Tang, Jiajin Li, Ziqi Gao, and Jia Li. 2022. Rethinking Graph Neural Net- works for Anomaly Detection. In Proceeding of the 39th International Conference on Machine Learning, Vol. 162

    2022

  35. [43]

    Jie Tang, Jing Zhang, Limin Yao, Juanzi Li, Li Zhang, and Zhong Su. 2008. Ar- netminer: extraction and mining of academic social networks. In Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining. 990–998

    2008

  36. [44]

    Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al. 2023. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 (2023)

    2023 arXiv

  37. [45]

    Herun Wan, Shangbin Feng, Zhaoxuan Tan, Heng Wang, Yulia Tsvetkov, and Minnan Luo. 2024. DELL: Generating Reactions and Explanations for LLM- Based Misinformation Detection. In Findings of the Association for Computational Linguistics

    2024

  38. [46]

    Tingmin Wu, Sheng Wen, Yang Xiang, and Wanlei Zhou. 2018. Twitter spam detection: Survey of new approaches and comparative study. Computers and Security 76 (2018)

    2018

  39. [47]

    Qiang Yan and AsirAsir. 2024. Synergizing large language models and tree-based algorithms for author name disambiguation. In KDD 2024 OAG-Challenge Cup

    2024

  40. [48]

    An Yang, Baosong Yang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Zhou, Cheng- peng Li, Chengyuan Li, Dayiheng Liu, Fei Huang, et al. 2024. Qwen2 technical report. arXiv preprint arXiv:2407.10671 (2024)

    2024 arXiv

  41. [49]

    An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, and et al. 2024. Qwen2.5 Technical Report. CoRR abs/2412.15115 (2024). GuARD: Effective Anomaly Detection through a Text-Rich and Graph-Informed Language Model KDD ’25, August 3–7, 2025, Toronto, ON, Canada

    2024 arXiv

  42. [50]

    Rossi, Kaize Ding, Xiaohu You, and Yue Zhao

    Tiankai Yang, Yi Nian, Shawn Li, Ruiyao Xu, Yuangang Li, Jiaqi Li, Zhuo Xiao, Xiyang Hu, Ryan A. Rossi, Kaize Ding, Xiaohu You, and Yue Zhao. 2024. AD- LLM: Benchmarking Large Language Models for Anomaly Detection. CoRR abs/2412.11142 (2024)

    2024

  43. [51]

    Aohan Zeng, Xiao Liu, Zhengxiao Du, Zihan Wang, Hanyu Lai, Ming Ding, Zhuoyi Yang, Yifan Xu, Wendi Zheng, Xiao Xia, et al. 2022. Glm-130b: An open bilingual pre-trained model. Proceedings of the 5th International Conference on Learning Representations (2022)

    2022

  44. [52]

    Fanjin Zhang, Xiao Liu, Jie Tang, Yuxiao Dong, Peiran Yao, Jie Zhang, Xiaotao Gu, Yan Wang, Evgeny Kharlamov, Bin Shao, et al. 2023. OAG: Linking Entities across Large-scale Heterogeneous Knowledge Graphs. IEEE Transactions on Knowledge and Data Engineering 35, 9 (2023), 9225–9239

    2023

  45. [53]

    Fanjin Zhang, Xiao Liu, Jie Tang, Yuxiao Dong, Peiran Yao, Jie Zhang, Xiaotao Gu, Yan Wang, Bin Shao, Rui Li, et al. 2019. OAG: Toward linking large-scale heterogeneous entity graphs. InProceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data...

    2019

  46. [54]

    Fanjin Zhang, Shijie Shi, Yifan Zhu, Bo Chen, Yukuo Cen, Jifan Yu, Yelin Chen, Lulu Wang, Qingfei Zhao, Yuqing Cheng, et al . 2024. OAG-Bench: A human- curated benchmark for academic graph mining. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data...

    2024

  47. [55]

    Xiaocheng Zhang, Yang Zhou, Haoru Chen, Mengjiao Bao, and Peng Yan. 2024. Enhanced name disambiguation via iterative self-refining with LLMs. In KDD 2024 OAG-Challenge Cup

    2024

  48. [56]

    Wayne Xin Zhao, Kun Zhou, Junyi Li, Tianyi Tang, Xiaolei Wang, Yupeng Hou, Yingqian Min, Beichen Zhang, Junjie Zhang, Zican Dong, et al. 2023. A survey of large language models. arXiv preprint arXiv:2303.18223 (2023)

    2023 arXiv

  49. [57]

    +sem” and “+graph

    Wei Zhuo, Zemin Liu, Bryan Hooi, Bingsheng He, Guang Tan, Rizal Fathony, and Jia Chen. 2024. Partitioning Message Passing for Graph Fraud Detection. In Proceeding of the 12th International Conference on Learning Representations . A Implementation Details During the training of...

    2024

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.