Pith. sign in

REVIEW 4 major objections 6 minor 54 references

Integrate Temporal Graph Learning into LLM-based Temporal Knowledge Graph Model

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read TGL-LLM claims that an LLM forecasts events in a temporal knowledge graph far better when it receives the last few timestamps' entity embeddings as a time-ordered token sequence than when it receives retrieved text or a static graph…

desk verdict A well-engineered embedding-based LLM method with a genuinely new combination, but the headline margin rests on a non-standard multiple-choice protocol and has not been shown to survive full-entity ranking. read the letter →

arxiv 2501.11911 v1 pith:OROQ7FNR submitted 2025-01-21 cs.IR

He Chang , Jie Wu , Zhulin Tao , Yunshan Ma , Xianglin Huang , Tat-Seng Chua This is my paper · ORCID
classification cs.IR
keywords TemporalKnowledgeGraphForecastingLargeLanguageModelsHybridTokenizationLearningInfluenceFunctionDataPruningLoRAFine-tuningPOLECATDatasetMultiple-ChoiceEvent
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

TGL-LLM sets out to show that the bottleneck in LLM-based temporal knowledge graph forecasting is the form in which graph history is presented: retrieved text or a static graph embedding hides the order in which events unfold. The paper's claim is that encoding the most recent timestamps as entity embeddings and injecting them into the LLM as a time-ordered token sequence lets the language model exploit temporal patterns it otherwise cannot see. On three POLECAT subsets the authors report Acc@4 of 0.8514, 0.8779, and 0.8109, against a best baseline of 0.6083, 0.6110, and 0.5315. If this holds, the path to better LLM forecasting is to model time explicitly in the prompt rather than to add more textual history.

What carries the argument

The load-bearing object is the hybrid prompt: a time-ordered sequence of recent historical entity embeddings, one token per snapshot, projected into the LLM's token space by an entity adapter and separated by a learned feature token. The same construction describes the query subject, the query relation, and each candidate answer, so the LLM reasons over an evolving entity state rather than a frozen snapshot or a textual list of past facts.

What would settle it

Two probes would settle it: replace the random-recent-negative MCQ protocol with full-entity ranking or with near-miss negatives, and shuffle the order of the temporal graph tokens; if accuracy collapses in either case, the mechanism claimed is not the one driving the results.

Watch

Extended reading notes

Core claim

The framework, TGL-LLM, learns an entity's recent history with a temporal graph model: an RGCN aggregates each timestamp's relational neighborhood, and during training a GRU passes over the snapshots while a ConvTransE decoder with cross-entropy loss fits the graph model. For the LLM input, the framework keeps the recent T RGCN embeddings rather than the recurrent summary, and projects each snapshot into token space with small entity and relation adapters. A learned feature token marks the start of each graph segment, and the tokens appear in chronological order as the description of the subject, relation, and candidate answers. A two-stage LoRA fine-tune, first on influence-function-selected high-quality data and then on a smaller diversity subset, aligns this graph modality with language. The paper reports that this combination outperforms every Non-LLM and LLM-based baseline on all three POLECAT subsets by a large margin.

Load-bearing premise

The evaluation turns forecasting into a multiple-choice test whose wrong answers are random recent entities, so the reported accuracy may overstate how well the model would rank the true answer against harder near-miss candidates or the full entity set.

Editorial extensions

If this is right

  • Temporal graph tokens make the prompt shorter than text-based retrieval: TGL-LLM uses about 245 tokens per query versus 466 for CoH, while reaching higher accuracy, so the gain does not come from a larger context window.
  • Historical length is a real hyperparameter: accuracy rises from one day to about five to seven days of snapshots and then converges, which the paper reads as longer histories adding noise rather than signal.
  • The two-stage training data matters: removing either the influence-selected high-quality subset or the diversity subset lowers accuracy in most settings, showing that data quality and diversity both contribute to graph-language alignment.
  • On long-tail entities with few training links, TGL-LLM keeps a much smaller accuracy gap than non-LLM baselines, supporting the paper's argument that LLM knowledge plus aligned graph tokens handles sparse entities better.
  • Because the test window lies after the Llama2 training cutoff, the reported advantage is less likely to come from the LLM having memorized the test events.

Reading between the lines

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

  • Editorial inference: the random-negatives-from-the-last-three-days MCQ protocol may be an easier test than full-entity ranking; near-miss negatives with similar recent activity would be a harder probe, and full ranking would show whether the margin survives.
  • Editorial inference: the ablation results suggest the ordered recent snapshots themselves, not the GRU summary, may be what the LLM exploits; a probe that shuffles the token order would isolate temporal order as the active ingredient.
  • Editorial inference: the influence-plus-diversity data recipe is not tied to TKGF and could be transferred to other graph-to-text LLM alignments, such as static knowledge graph completion or recommendation, with the same adapter design.
  • Editorial inference: on POLECAT-IS the diversity stage adds little, so the optimal data recipe may vary by dataset; a cheaper single-stage variant could match performance when test patterns resemble training patterns.
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

4 major / 6 minor

Summary. The paper proposes TGL-LLM, an LLM-based framework for temporal knowledge graph forecasting. It pre-trains a temporal graph model (RGCN with GRU and ConvTransE) to obtain historical entity and relation embeddings, converts the recent T steps of these embeddings into language tokens through entity/relation adapters and a learned feature token, and fine-tunes Llama2-7B with LoRA in two stages: first on a high-quality subset selected by an influence function, then on a smaller diversity subset. Experiments on three POLECAT subsets (Iran, Israel, Egypt) under a multiple-choice evaluation protocol report large gains over Non-LLM and LLM-based baselines, with ablations showing contributions from hybrid graph tokenization and, to a smaller degree, the two-stage training paradigm.

Significance. If the reported results hold, the paper's core idea of feeding temporally ordered graph tokens into an LLM is a plausible and useful direction for LLM-based TKGF, and the choice of POLECAT data after the LLM training cutoff is a genuine strength for avoiding leakage. However, the central SOTA claim currently rests on a non-standard candidate-construction protocol, and the statistical evidence is weak. The paper does not release code or data, does not report error bars, and the ANOVA significance claim is not supported by the presented experimental detail. The contribution is therefore promising but not yet convincingly established; the evaluation protocol and experimental rigor need substantial work before the claims can be accepted.

major comments (4)
  1. [§5.1.2, Table 2] The central claim that TGL-LLM outperforms all baselines by a large margin rests on a multiple-choice protocol in which the 4/6/10 candidate entities are randomly sampled from the most recent three days, always including the positive entity. This is not the standard TKGF evaluation (full-entity ranking with time-aware filtering), so the Acc@4/6/10 numbers in Table 2 are not comparable to Hits@K/MRR reported in prior work. More importantly, the protocol may systematically favor TGL-LLM because its hybrid prompt feeds recent graph tokens of each candidate to the LLM, turning the task into a recency/co-occurrence matching problem rather than open forecasting. The justification in Appendix A.3 studies GenTKG on ICEWS/GDELT and only shows that generated answers usually appear in the prompt; it does not validate that recent-three-day negatives are representative for POLECAT or that the conclusion transfers to TGL-LLM. Please add a full-entity ranking evaluation (or at least negatives sampled from the full entity set or from a longer history) and report Hits@K/MRR; without this, the 'large margin' conclusion is not established.
  2. [§5.1.4, §4.3.2] The training-data budget appears inconsistent across methods. Section 5.1.4 says 'each experiment is trained on 100,000 training samples randomly selected from the training dataset,' but the same paragraph sets the high-quality subset |D_h| to 100,000 and the diversity subset |D_p| to 10,000. If TGL-LLM is fine-tuned on 100,000 + 10,000 = 110,000 samples while KoPA, CoH, and GenTKG receive 100,000 samples, the comparison in Table 2 is not matched in training data. Please clarify how the random 100,000-sample set relates to D_h and D_p, and either match the total number of training samples or updates across all methods or add a baseline trained on 110,000 samples.
  3. [§5.2, Tables 2–4] The paper states that 'ANOVA significance test [over] the performances of all baselines. All of the evaluated p-values are below 0.01,' but Table 2 contains a single performance number per condition and no variance estimate or number of repeated runs. ANOVA requires within-condition variability; none is reported. The same problem affects Tables 3 and 4, where several differences are small (e.g., Acc@4 on POLECAT-IR: Random 0.8467 vs. TGL-LLM 0.8514; w/o-DS 0.8354). Please report means and standard deviations over at least three seeds and use an appropriate significance test, or remove the ANOVA claim.
  4. [§4.3.1, Eq. (11)] The data-pruning component, one of the two key innovations, is under-specified. Equation (11) defines influence scores, but the text does not describe how the scores are converted into the high-quality subset D_h: what 'stratified sampling' stratifies on, what threshold or budget is used, and how the 100,000-sample size is chosen. Since the surrogate graph model is trained on the same training set, the influence scores measure fit to the graph model's training loss, not necessarily informativeness for LLM cross-modal alignment; the assertion that high influence indicates high quality for LLM fine-tuning needs a more direct justification or an experiment showing that the selected subset is not simply the graph model's easy samples. Please provide the full selection algorithm and implementation details.
minor comments (6)
  1. [Table 3] The POLECAT-IS Acc@10 value for ConvTransE is written as '0,6033' instead of '0.6033'.
  2. [Figure 3] The legend and axis contain the typo 'Saprse'; it should be 'Sparse'.
  3. [§4.3.1, §3, §5.4.1] There are several typos: 'issuse' should be 'issues', 'classicial' should be 'classical', and 'CovTransE' should be 'ConvTransE'.
  4. [Appendix A.3] The statistical analysis uses GenTKG on ICEWS/GDELT, not the POLECAT subsets or TGL-LLM. Please clarify how these results support the multiple-choice protocol for the datasets and model used in this paper, or soften the equivalence claim.
  5. [§5.4.2, Figure 4] The historical-length study reports a single curve per dataset without error bars, so the conclusions about the optimal history length (5 vs. 7) may reflect noise; please add variance estimates.
  6. [Reference [47]] The KoPA citation appears with year 2021; please verify that this is the correct publication year and venue for the cited work.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: TGL-LLM is trained on historical data and evaluated on future test events; self-citations are not load-bearing.

full rationale

The claimed derivation chain is self-contained. The temporal graph model (RGCN+GRU+ConvTransE) is trained on the training split only, using a next-timestamp link prediction loss (Eq. 4) that depends only on historical graphs within the training period; the LLM is fine-tuned on the same historical training split via LoRA (Eq. 12). Test events are from November 2023 to April 2024, after Llama2's cutoff, and no test labels enter the influence-function data pruning (Eqs. 9-11), which scores samples by their effect on the graph model's training loss. The MCQ evaluation protocol, which constructs candidate sets by random sampling from the most recent three days and always includes the positive entity, is a benchmarking choice that may affect accuracy and deserves correctness scrutiny, but it does not make the reported Acc@K equal to an input by construction: the model must still select the correct entity from the candidates, and the same protocol is applied to all baselines. The self-citation [5] supports only a background claim about long-tail entities and is not load-bearing; no uniqueness theorem or ansatz is imported from self-citation. The central SOTA claim is therefore empirically testable and not circular.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The paper is an empirical ML work with no formal derivation. It relies on several unproven assumptions: the equivalence of MCQ and generative TKGF, the usefulness of RGCN+GRU embeddings after projection, and the transferability of influence scores from a graph surrogate to LLM fine-tuning. It also requires selecting hyperparameters such as history length and subset sizes, which are tuned per dataset.

free parameters (6)
  • history length T = 5 for POLECAT-IR/IS, 7 for POLECAT-EG
    Selected by validation-set search (Figure 4); central to hybrid graph tokenization.
  • number of RGCN layers L = 2
    Set in implementation details (Section 5.1.4); affects graph embedding depth.
  • embedding size d = 200
    Set for graph model and LLM adapters (Section 5.1.4).
  • high-quality subset size |D_h| = 100,000
    Used for Stage 1 finetuning; chosen by authors without ablating size (Section 5.1.4).
  • diversity subset size |D_p| = 10,000
    Used for Stage 2; authors state it is 'significantly smaller' but do not justify exact value (Section 5.1.4).
  • negatives sampling window = 3 days
    Candidates drawn from entities in the most recent three days (Section 5.1.2); not ablated and could influence difficulty.
assumptions (4)
  • domain assumption The MCQ setting is a faithful and equivalent approximation of generative TKGF for LLMs.
    Argued in Section 3 and Appendix A.3 using GenTKG statistics; if false, the reported absolute accuracies do not transfer to standard TKGF.
  • domain assumption RGCN+GRU pre-trained on historical data yields graph embeddings that, after projection, provide useful temporal context to the LLM.
    Core to hybrid tokenization; supported only by ablations on the same datasets, not by external benchmarks.
  • ad hoc to paper Influence scores computed with the graph surrogate model identify samples that improve LLM fine-tuning.
    Used for data pruning in Section 4.3.1; the surrogate differs from the fine-tuned LLM, so transferability is assumed.
  • domain assumption Test and part of validation data are outside Llama2's pretraining knowledge due to the July 2023 cutoff.
    Section 5.1.1; only a timeline argument is given, no contamination probe is performed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Integrate Temporal Graph Learning into LLM-based Temporal Knowledge Graph Model." pith.science (2026). https://pith.science/paper/OROQ7FNR

@misc{pith2026250111911,
  author       = {Pith},
  title        = {Pith review of: Integrate Temporal Graph Learning into LLM-based Temporal Knowledge Graph Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OROQ7FNR}},
  note         = {Machine review of arXiv:2501.11911}
}
read the original abstract

Temporal Knowledge Graph Forecasting (TKGF) aims to predict future events based on the observed events in history. Recently, Large Language Models (LLMs) have exhibited remarkable capabilities, generating significant research interest in their application for reasoning over temporal knowledge graphs (TKGs). Existing LLM-based methods have integrated retrieved historical facts or static graph representations into LLMs. Despite the notable performance of LLM-based methods, they are limited by the insufficient modeling of temporal patterns and ineffective cross-modal alignment between graph and language, hindering the ability of LLMs to fully grasp the temporal and structural information in TKGs. To tackle these issues, we propose a novel framework TGL-LLM to integrate temporal graph learning into LLM-based temporal knowledge graph model. Specifically, we introduce temporal graph learning to capture the temporal and relational patterns and obtain the historical graph embedding. Furthermore, we design a hybrid graph tokenization to sufficiently model the temporal patterns within LLMs. To achieve better alignment between graph and language, we employ a two-stage training paradigm to finetune LLMs on high-quality and diverse data, thereby resulting in better performance. Extensive experiments on three real-world datasets show that our approach outperforms a range of state-of-the-art (SOTA) methods.

Figures

Figures reproduced from arXiv: 2501.11911 by the authors.

Figure 1
Figure 1. Illustration of the limitations in the current [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overall framework of TGL-LLM consists of two key novel modules: 1) hybrid graph tokenization that sufficiently [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 4
Figure 4. Performance comparison of our method when using [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figures from the paper (6 more)
Figure 3
Figure 3. Figure 3: The performance comparison across two datasets [PITH_FULL_IMAGE:figures/full_fig_p008_3.png]
Figure 5
Figure 5. Figure 5: Specific prompt used for entity filtering. [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: Heatmap illustration of event distribution by year [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: Heatmap illustration of article distribution by year [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: The top-20 most frequent entities in the dataset. [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 10
Figure 10. Figure 10: The instruction utilized in the hybrid graph tok [PITH_FULL_IMAGE:figures/full_fig_p012_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 30 canonical work pages

  1. [1]

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Floren- cia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774 (2023)

    arXiv 2023

  2. [2]

    Naman Agarwal, Brian Bullins, and Elad Hazan. 2016. Second-order stochastic optimization in linear time. stat 1050 (2016), 15

    work page 2016

  3. [3]

    Luyi Bai, Wenting Yu, Die Chai, Wenjun Zhao, and Mingzhuo Chen. 2023. Tem- poral knowledge graphs reasoning with iterative guidance by temporal logical rules. Information Sciences 621 (2023), 22–35

    work page 2023

  4. [4]

    Li Cai, Xin Mao, Yuhao Zhou, Zhaoguang Long, Changxu Wu, and Man Lan

  5. [5]

    He Chang, Chenchen Ye, Zhulin Tao, Jie Wu, Zhengmao Yang, Yunshan Ma, Xianglin Huang, and Tat-Seng Chua. 2024. A Comprehensive Evaluation of Large Language Models on Temporal Event Forecasting. arXiv:2407.11638 [cs.CL] https://arxiv.org/abs/2407.11638

    arXiv 2024

  6. [6]

    Songgaojun Deng, Huzefa Rangwala, and Yue Ning. 2020. Dynamic Knowledge Graph based Multi-Event Forecasting. In KDD. ACM, 1585–1595

    work page 2020

  7. [7]

    Songgaojun Deng, Huzefa Rangwala, and Yue Ning. 2020. Dynamic knowledge graph based multi-event forecasting. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining . 1585–1595

    work page 2020

  8. [8]

    Songgaojun Deng, Huzefa Rangwala, and Yue Ning. 2021. Understanding Event Predictions via Contextualized Multilevel Feature Learning. In CIKM. ACM, 342– 351

    work page 2021

Show all 54 references
  1. [9]

    Zifeng Ding, Zongyue Li, Ruoxia Qi, Jingpei Wu, Bailan He, Yunpu Ma, Zhao Meng, Shuo Chen, Ruotong Liao, Zhen Han, et al. 2023. Forecasttkgquestions: a benchmark for temporal question answering and forecasting over temporal knowledge graphs. In International Semantic Web Confe...

    2023

  2. [10]

    Ellen R Girden. 1992. ANOV A: Repeated measures. Number 84. sage

    1992

  3. [11]

    Danny Halawi, Fred Zhang, Chen Yueh-Han, and Jacob Steinhardt. 2024. Ap- proaching Human-Level Forecasting with Language Models. arXiv preprint arXiv:2402.18563 (2024)

    2024 arXiv

  4. [12]

    Yupeng Hou, Junjie Zhang, Zihan Lin, Hongyu Lu, Ruobing Xie, Julian McAuley, and Wayne Xin Zhao. 2024. Large language models are zero-shot rankers for recommender systems. In European Conference on Information Retrieval. Springer, 364–381

    2024

  5. [13]

    Woojeong Jin, Meng Qu, Xisen Jin, and Xiang Ren. 2020. Recurrent Event Network: Autoregressive Structure Inferenceover Temporal Knowledge Graphs. In EMNLP (1). Association for Computational Linguistics, 6669–6683

    2020

  6. [14]

    Pang Wei Koh and Percy Liang. 2017. Understanding black-box predictions via influence functions. In International conference on machine learning . PMLR, 1885–1894

    2017

  7. [15]

    Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo, and Yusuke Iwasawa. 2022. Large Language Models are Zero-Shot Reasoners. In NeurIPS

    2022

  8. [16]

    Julien Leblay and Melisachew Wudage Chekol. 2018. Deriving validity time in knowledge graph. In Companion proceedings of the the web conference 2018 . 1771–1776

    2018

  9. [17]

    Dong-Ho Lee, Kian Ahrabian, Woojeong Jin, Fred Morstatter, and Jay Pujara. 2023. Temporal Knowledge Graph Forecasting Without Knowledge Using In-Context Learning. In EMNLP. Association for Computational Linguistics, 544–557

    2023

  10. [18]

    Dong-Ho Lee, Kian Ahrabian, Woojeong Jin, Fred Morstatter, and Jay Pujara

  11. [19]

    Kalev Leetaru and Philip A Schrodt. 2013. Gdelt: Global data on events, location, and tone, 1979–2012. In ISA annual convention, Vol. 2. Citeseer, 1–49

    2013

  12. [20]

    Haoxuan Li, Zhengmao Yang, Yunshan Ma, Yi Bin, Yang Yang, and Tat-Seng Chua

  13. [21]

    Zixuan Li, Zhongni Hou, Saiping Guan, Xiaolong Jin, Weihua Peng, Long Bai, Yajuan Lyu, Wei Li, Jiafeng Guo, and Xueqi Cheng. 2022. HiSMatch: Histor- ical Structure Matching based Temporal Knowledge Graph Reasoning. CoRR abs/2210.09708 (2022)

    2022 arXiv

  14. [22]

    Zixuan Li, Xiaolong Jin, Wei Li, Saiping Guan, Jiafeng Guo, Huawei Shen, Yuanzhuo Wang, and Xueqi Cheng. 2021. Temporal Knowledge Graph Rea- soning Based on Evolutional Representation Learning. In SIGIR. ACM, 408–417

    2021

  15. [23]

    arXiv preprint arXiv:2408.04388 (2024)

    MM-Forecast: A Multimodal Approach to Temporal Event Forecasting with Large Language Models. arXiv preprint arXiv:2408.04388 (2024)

    2024 arXiv

  16. [24]

    Xinyu Lin, Wenjie Wang, Yongqi Li, Shuo Yang, Fuli Feng, Yinwei Wei, and Tat- Seng Chua. 2024. Data-efficient Fine-tuning for LLM-based Recommendation. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval . 365–374

    2024

  17. [25]

    Robert F Ling. 1984. Residuals and influence in regression

    1984

  18. [26]

    Ruotong Liao, Xu Jia, Yunpu Ma, and Volker Tresp. 2023. GenTKG: Generative Forecasting on Temporal Knowledge Graph. CoRR abs/2310.07793 (2023)

    2023 arXiv

  19. [27]

    Ruilin Luo, Tianle Gu, Haoling Li, Junzhe Li, Zicheng Lin, Jiayi Li, and Yujiu Yang. 2024. Chain of History: Learning and Forecasting with LLMs for Temporal Knowledge Graph Completion. CoRR abs/2401.06072 (2024)

    2024 arXiv

  20. [28]

    Yunshan Ma, Chenchen Ye, Zijian Wu, Xiang Wang, Yixin Cao, and Tat-Seng Chua. 2023. Context-aware Event Forecasting via Graph Disentanglement. In KDD. ACM, 1643–1652

    2023

  21. [29]

    Yushan Liu, Yunpu Ma, Marcel Hildebrandt, Mitchell Joblin, and Volker Tresp

  22. [30]

    Farzaneh Mahdisoltani, Joanna Biega, and Fabian M Suchanek. 2013. Yago3: A knowledge base from multilingual wikipedias. In CIDR

    2013

  23. [31]

    Brando Miranda, Alycia Lee, Sudharsan Sundar, Allison Casasola, and Sanmi Koyejo. 2023. Beyond Scale: The Diversity Coefficient as a Data Quality Metric for Variability in Natural Language Data. arXiv e-prints (2023), arXiv–2306

    2023

  24. [32]

    Mehrnoosh Mirtaheri, Ryan A Rossi, Sungchul Kim, Kanak Mahadik, Tong Yu, Xiang Chen, and Mohammad Rostami. 2024. Tackling Long-Tail Entities for Temporal Knowledge Graph Completion. In Companion Proceedings of the ACM on Web Conference 2024. 497–500

    2024

  25. [33]

    Yunshan Ma, Chenchen Ye, Zijian Wu, Xiang Wang, Yixin Cao, Liang Pang, and Tat-Seng Chua. 2023. Structured, Complex and Time-complete Temporal Event Forecasting. CoRR abs/2312.01052 (2023)

    2023 arXiv

  26. [34]

    OpenAI. 2022. Introducing ChatGPT. https://openai.com/blog/chatgpt

    2022

  27. [35]

    Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul F. Christiano, Jan Leik...

    2022

  28. [36]

    Scarborough, Benjamin E

    Grace I. Scarborough, Benjamin E. Bagozzi, Andreas Beger, John Berrie, Andrew Halterman, Philip A. Schrodt, and Jevon Spivey. 2023. POLECAT Weekly Data. Harvard Dataverse. https://doi.org/10.7910/DVN/AJGVIT

    2023 doi

  29. [37]

    Sean P O’brien. 2010. Crisis early warning and decision support: Contemporary approaches and thoughts on future research. International studies review 12, 1 (2010), 87–104

    2010

  30. [38]

    Chao Shang, Yun Tang, Jing Huang, Jinbo Bi, Xiaodong He, and Bowen Zhou

  31. [39]

    Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aurélien Rodriguez, Armand Joulin, Edouard Grave, and Guillaume Lam- ple. 2023. LLaMA: Open and Efficient Foundation ...

    2023 arXiv

  32. [40]

    Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yas- mine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhos- ale, Dan Bikel, Lukas Blecher, Cristian Canton Ferrer, Moya Chen, Guillem Cucu- rull, David Esiobu, Jude Fernandes, Jeremy...

    2017 arXiv

  33. [41]

    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 Con- volutional Networks. In ESWC (Lecture Notes in Computer Science, Vol. 10843) . Springer, 593–607

    2018

  34. [42]

    Siheng Xiong, Yuan Yang, Faramarz Fekri, and James Clayton Kerce. 2022. TILP: Differentiable Learning of Temporal Logical Rules on Knowledge Graphs. In The Eleventh International Conference on Learning Representations

    2022

  35. [43]

    Siheng Xiong, Yuan Yang, Ali Payani, James C Kerce, and Faramarz Fekri. 2024. Teilp: Time prediction over knowledge graphs via logical reasoning. InProceedings of the AAAI Conference on Artificial Intelligence , Vol. 38. 16112–16119

    2024

  36. [44]

    Wenjie Xu, Ben Liu, Miao Peng, Xu Jia, and Min Peng. 2023. Pre-trained Lan- guage Model with Prompts for Temporal Knowledge Graph Completion. In ACL (Findings). Association for Computational Linguistics, 7790–7803

    2023

  37. [45]

    Bishan Yang, Wen-tau Yih, Xiaodong He, Jianfeng Gao, and Li Deng. 2015. Em- bedding Entities and Relations for Learning and Inference in Knowledge Bases. In ICLR (Poster)

    2015

  38. [46]

    Duo Wang, Yuan Zuo, Fengzhi Li, and Junjie Wu. 2024. LLMs as Zero-shot Graph Learners: Alignment of GNN Representations with LLM Token Embeddings. arXiv preprint arXiv:2408.14512 (2024)

    2024 arXiv

  39. [47]

    Yichi Zhang, Zhuo Chen, Wen Zhang, and Huajun Chen. 2021. Making large language models perform better in knowledge graph completion. In MM. ACM, 342–351

    2021

  40. [48]

    Liang Zhao. 2022. Event Prediction in the Big Data Era: A Systematic Survey. ACM Comput. Surv. 54, 5 (2022), 94:1–94:37

    2022

  41. [49]

    yes" or

    Haizhong Zheng, Rui Liu, Fan Lai, and Atul Prakash. 2022. Coverage-centric coreset selection for high pruning rates. arXiv preprint arXiv:2210.15809 (2022). A APPENDIX A.1 Details on Dataset Construction This section presents the sources and steps involved in constructing a ne...

    2022 arXiv

  42. [51]

    Mengmei Zhang, Mingwei Sun, Peng Wang, Shen Fan, Yanhu Mo, Xiaoxiao Xu, Hong Liu, Cheng Yang, and Chuan Shi. 2024. GraphTranslator: Aligning Graph Model to Large Language Model for Open-ended Tasks. In Proceedings of the ACM on Web Conference 2024. 1003–1014

    2024

  43. [2019]

    End-to-End Structure-Aware Convolutional Networks for Knowledge Base Completion. In AAAI. AAAI Press, 3060–3067

  44. [2022]

    In Proceedings of the AAAI conference on artificial intelligence , Vol

    Tlogic: Temporal logical rules for explainable link forecasting on temporal knowledge graphs. In Proceedings of the AAAI conference on artificial intelligence , Vol. 36. 4120–4127

  45. [2023]

    arXiv preprint arXiv:2305.10613 (2023)

    Temporal knowledge graph forecasting without knowledge using in-context learning. arXiv preprint arXiv:2305.10613 (2023)

    2023 arXiv

  46. [2024]

    arXiv preprint arXiv:2403.04782 (2024)

    A Survey on Temporal Knowledge Graph: Representation Learning and Applications. arXiv preprint arXiv:2403.04782 (2024)

    2024 arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.