Pith. sign in

REVIEW 3 major objections 6 minor 104 references

A Survey on LLM-based News Recommender Systems

T0 review · 3 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This survey claims to be the first systematic, experiment-backed review of LLM-based news recommender systems, and its benchmark shows LLM-based models beat deep learning baselines on MIND but not on Adressa.

desk verdict Useful survey of LLM-based news recommendation, but the benchmark's degenerate baselines make its quantitative conclusions unsupported. read the letter →

arxiv 2502.09797 v2 pith:2CWJ24OL submitted 2025-02-13 cs.IR cs.AI

Rongyao Wang , Veronica Liesaputra , Zhiyi Huang This is my paper · ORCID
classification cs.IRcs.AI
keywords newsrecommendersystemslargelanguagemodelsdiscriminativeLLMsgenerativebenchmarkevaluationMINDdatasetAdressadiversityandpersonalization
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

This paper claims to be the first systematic survey of news recommender systems in the LLM era that combines a literature review with original benchmark experiments. It sorts LLM-based recommenders into discriminative and generative families, and organizes their methods by what they model: news content, user profiles, or the prediction step. The benchmark on MIND-small and one-week Adressa finds that LLM-based models can beat deep learning baselines on the English MIND data, with MANNeR reaching the best AUC and LKPNR the best recall, but that deep learning models such as CAUM and LSTUR remain ahead on Adressa. The paper concludes that LLM-based recommenders are promising but not uniformly superior, and that their gains depend on dataset scale and language.

What carries the argument

The central organizing device is a unified framework that splits LLM-based news recommenders into news-oriented, user-oriented, and prediction-oriented modeling, overlaid on the DLLM/GLLM distinction. Discriminative LLMs are pretrained encoders such as BERT that produce text representations; generative LLMs such as GPT-4 and LLaMA create or rewrite content. The load-bearing experimental machinery is a common benchmark harness that trains a set of deep learning, DLLM-based, and GLLM-based models on MIND-small and one-week Adressa and evaluates them on classification, ranking, diversity, and personalization metrics, with five repeated runs for mean and standard deviation.

What would settle it

Re-run the DLLM-empowered baselines (LSTUR-DLLM, TANR-DLLM, NRMS-DLLM, NAML-DLLM) on MIND-small from official code: if their AUC moves clearly away from 50.0, the paper's conclusion that DLLM-empowered models underperform at small scale is an artifact of broken baselines. A second check would fine-tune a Norwegian-capable encoder and re-run Adressa to test whether the language explanation, rather than model capacity, drives the gap.

Watch

Extended reading notes

Core claim

The central claim is that a systematic, experiment-backed review changes how the field should read LLM-based news recommendation: these methods are best understood through three modeling aspects, news-oriented, user-oriented, and prediction-oriented, and their reported superiority over deep learning is real but conditional. The paper's own experiments show that on MIND-small, the DLLM-based MANNeR exceeds the best deep baseline by about 13 percent in AUC and the GLLM-based LKPNR beats NAML by about 7 percent in recall, while on Norwegian Adressa the deep models CAUM and LSTUR beat all LLM-based models. DLLM-empowered versions of LSTUR, TANR, NRMS, and NAML do not outperform their original deep-learning versions in this small-scale setup, which the authors attribute to insufficient training data for the pretrained encoders. The experimental section therefore supports a hedged conclusion: LLM-based recommenders show strong potential in favorable settings, but deep learning remains competitive or better in others.

Load-bearing premise

The experimental comparison rests on the assumption that every reproduced model was trained and evaluated correctly with comparable settings, yet rows reporting exactly 50.0 AUC with zero standard deviation, such as LSTUR-DLLM and NRMS-DLLM on MIND, suggest some reproductions may have failed rather than genuinely underperformed.

Editorial extensions

If this is right

  • On English datasets at MIND-small scale, LLM-based recommenders can deliver large gains: MANNeR raises AUC by about 13 percent over TANR and LKPNR raises recall by about 7 percent over NAML.
  • These gains do not transfer to Norwegian data: CAUM and LSTUR beat every LLM-based model on Adressa, so practitioners should validate LLM recommenders per language.
  • Swapping a pretrained encoder into an existing deep model does not guarantee improvement at small scale; the DLLM-empowered variants trail their originals in these experiments.
  • GLLM-based models are not automatically better than DLLM-based models; LKPNR trails MANNeR on AUC while beating it on recall.
  • Diversity and personalization results are mixed, with LLM models improving category diversity on MIND while deep models like LSTUR keep personalization advantages.

Reading between the lines

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

  • The exactly-50.0 AUC rows are a red flag for the benchmark's internal validity; a corrected rerun could remove the small-scale DLLM underperformance result or explain it in terms of training instability.
  • The Adressa gap is consistent with English-centric pretraining rather than a fundamental limit of LLM encoders; fine-tuning on Norwegian or using multilingual models is the natural test.
  • The survey's one-week Adressa and MIND-small scale is too small to settle the broader claim that LLM recommenders outperform deep learning; MIND-large and EB-NeRD would be stronger evidence.
  • The three-aspect taxonomy could be applied to the social-driven and multimodal directions the paper lists, making the framework a reusable lens rather than just a review device.
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

3 major / 6 minor

Summary. This manuscript presents a survey of LLM-based news recommender systems, organizing methods into DLLM- and GLLM-based categories and reviewing them along three axes: news-oriented, user-oriented, and prediction-oriented modeling. It also discusses datasets, benchmarking tools, methodological challenges, and future directions. A central contribution claimed by the authors is the first systematic survey that includes extensive comparative experiments: they benchmark nine deep learning models, six DLLM-based models, and two GLLM-based models on MIND-small and one-week Adressa, measuring classification, ranking, diversity, and personalization metrics, and use these results to answer four research questions (Q1-Q4). The survey's taxonomy and literature coverage are plausible, but the experimental evidence, which underlies several headline conclusions, contains degenerate baseline results and incomplete provenance, making the comparative findings unreliable in their current form.

Significance. If the experimental results were sound, the paper would be a useful contribution: it provides a structured organization of recent LLM-based news recommendation research, a comparison of benchmarking tools, and a rare head-to-head evaluation of deep learning, DLLM, and GLLM methods on two popular datasets. The authors also deserve credit for going beyond a purely narrative survey and for attempting to measure diversity and personalization, which are often ignored. However, the significance is currently conditional: the degenerate AUC values and the unclear provenance of the reported results prevent the experimental conclusions from being accepted as evidence. The survey portion alone may still be publishable, but the paper's stated novelty—'the first systematic survey ... by conducting extensive experiments'—hinges on the reliability of Section V.

major comments (3)
  1. [§V.E.3 and Table IV] Table IV reports AUC = 50.00 with standard deviation 0.00 for LSTUR-DLLM and NRMS-DLLM on MIND, and for NAML and MANNeR on Adressa, with additional near-50 values for other models. A standard deviation of exactly zero over five independent runs is not a signature of an undertrained but stochastic model; it indicates a degenerate predictor, such as a constant score or an evaluation pipeline that never sees informative gradients. The explanation in §V.E.3 that 'DLLMs are not effectively trained on a small-scale dataset' is not credible for exactly 50.00 with zero variance. Since Q3's conclusion ('DLLM-empowered models do not perform superior to the original models') relies directly on these rows, and Q1 and Q4 compare against these baselines, the experimental findings are not supported. Please re-run these models with a verified training/evaluation procedure, report corrected numbers, or explicitly exclude broken baselines and weaken the corresponding conclusions.
  2. [§V.D] The experimental setup contains a major provenance inconsistency. The text says 'We use similar configurations with NewsRecLib [71] in terms of deep learning-based news recommendation models and DLLM-based news recommendation models. Besides, we employ official codes to reproduce GLLM-based news recommendation models (e.g., ONCE, LKPNR) on the MIND dataset. Specifically, we only reproduce DIRE (Discriminative Recommendation Framework)-NAML to conduct our main experiments in Table IV.' This leaves it unclear whether any row in Table IV was actually run by the authors beyond DIRE-NAML, and what 'DIRE-NAML' refers to. For a benchmark-based survey, the authors should provide a per-model provenance table listing which rows were freshly reproduced, which use library outputs, and which are taken from prior publications, together with hyperparameters, training epochs, and hardware details.
  3. [§V.B and Tables V-VI] The diversity and personalization metrics are under-specified. In Equation (8), p(j) is never defined; presumably it is the proportion of recommended items with aspect j, but this must be stated, and the exact mapping to 'categ div' and 'sent div' is not given. Equation (9) defines P SAp@k, yet the relation between this formula and the reported 'categ pers' and 'sent pers' columns is not explained, and it is unclear how sentiment labels are obtained for Adressa. Without these definitions, the Q2 results cannot be interpreted or reproduced.
minor comments (6)
  1. [Table IV caption] The caption says 'Gary's color represents LLM-based news recommendation models'; this should be 'Gray's color' or 'Grey's color'.
  2. [§III.A] There is a typo: 'technolegies' should be 'technologies'.
  3. [Table II] The formatting of the Plista row is unclear: the entries under '# News' and '# User' contain slashes (e.g., '4 70,353 / 1,095,323'), and the '# Behavior' column for Plista is empty. Please align the table entries with the dataset descriptions in Section IV.A.
  4. [§VII] The conclusion states that 'In the future, we will evaluate the diversity and personalization of LLM-based news recommendation models', but Tables V and VI already report such evaluations. This is contradictory and should be corrected to refer to a different future direction, such as evaluating GLLM models on diversity and personalization.
  5. [§V.D] The abbreviation 'DIRE (Discriminative Recommendation Framework)-NAML' is not defined or referenced. If DIRE is a framework proposed elsewhere, a citation is needed; otherwise the term should be clarified or replaced.
  6. [Equations (3)-(4)] Equations (3) and (4) lack the multiplication signs (e.g., 'TP F P' should be 'TP + FP'), making the formulas hard to read. Please typeset them properly.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the survey's organization and experimental benchmark are independent of its reviewed literature and of the authors' prior work.

full rationale

This paper is a survey with an appended benchmark study; it does not derive results from definitions or fit parameters to data in a way that could be circular. The central contribution claim—being the first systematic survey with extensive experiments—is a novelty claim, not a derivation, and the experimental conclusions in Section V are read off tables of measured metrics rather than obtained by construction. The authors do cite their own prior work (e.g., [8], [12], [53] by Rongyao Wang), but those citations are part of the reviewed literature and do not support the survey's organization or its experimental findings. The benchmark uses NewsRecLib [71] and official code for GLLM models, which are external artifacts, so the Q1–Q4 answers are not forced by a self-citation chain. The passage 'we only reproduce DIRE (Discriminative Recommendation Framework)-NAML to conduct our main experiments in Table IV' and the presence of exactly-50.0 AUC entries with zero standard deviation in Table IV raise reproducibility and correctness concerns, but those are empirical validity issues, not circular reasoning: the paper does not define the DLLM baselines in terms of the conclusion that they underperform. Under the hard rules, concern about broken baselines belongs under correctness risk, not circularity, and no quoted equation or self-citation reduces the paper's claims to its own inputs. The honest finding is therefore no significant circularity.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

The paper introduces no free parameters or invented entities. The central claims rest on standard evaluation protocols and the assumption that the baseline implementations and small dataset subsets are representative. No novel derived quantity is fitted.

assumptions (2)
  • domain assumption The implementations from NewsRecLib and official repositories accurately represent the original news recommendation models.
    Section V.D states the authors use NewsRecLib and official codes (ONCE, LKPNR) for the benchmark, so the validity of the comparison depends on these implementations being faithful.
  • domain assumption MIND-small and one-week Adressa are sufficient to draw conclusions about LLM-based versus deep learning-based news recommender performance.
    Section V.A limits experiments to these subsets, and the conclusions in Section V.E generalize from them without external validation on other datasets.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Survey on LLM-based News Recommender Systems." pith.science (2026). https://pith.science/paper/2CWJ24OL

@misc{pith2026250209797,
  author       = {Pith},
  title        = {Pith review of: A Survey on LLM-based News Recommender Systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2CWJ24OL}},
  note         = {Machine review of arXiv:2502.09797}
}
read the original abstract

News recommender systems play a critical role in mitigating the information overload problem. In recent years, due to the successful applications of large language model technologies, researchers have utilized Discriminative Large Language Models (DLLMs) or Generative Large Language Models (GLLMs) to improve the performance of news recommender systems. Although several recent surveys review significant challenges for deep learning-based news recommender systems, such as fairness, privacy-preserving, and responsibility, there is a lack of a systematic survey on Large Language Model (LLM)-based news recommender systems. In order to review different core methodologies and explore potential issues systematically, we categorize DLLM-based and GLLM-based news recommender systems under the umbrella of LLM-based news recommender systems. In this survey, we first overview the development of deep learning-based news recommender systems. Then, we review LLM-based news recommender systems based on three aspects: news-oriented modeling, user-oriented modeling, and prediction-oriented modeling. Next, we examine the challenges from various perspectives, including datasets, benchmarking tools, and methodologies. Furthermore, we conduct extensive experiments to analyze how large language model technologies affect the performance of different news recommender systems. Finally, we comprehensively explore the future directions for LLM-based news recommendations in the era of LLMs.

Figures

Figures reproduced from arXiv: 2502.09797 by the authors.

Figure 1
Figure 1. The tendency of news recommendation papers [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. A general uniform news recommendation frame [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The Hit Rate on the MIND and Adressa dataset. [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: The Precision on the MIND and Adressa dataset. [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

104 extracted references · 68 canonical work pages

  1. [71]

    Newsreclib: A pytorch-lightning library for neural news recommendation,

    ——, “Newsreclib: A pytorch-lightning library for neural news recommendation,” arXiv preprint arXiv:2310.01146 , 2023

    arXiv 2023

  2. [1]

    NPA: Neural news recommendation with personalized attention,

    C. Wu, F. Wu, M. An, J. Huang, Y . Huang, and X. Xie, “NPA: Neural news recommendation with personalized attention,” in KDD, 2019, pp. 2576–2584. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 17

    2019

  3. [2]

    Drn: A deep reinforcement learning framework for news recommendation,

    G. Zheng, F. Zhang, Z. Zheng, Y . Xiang, N. J. Yuan, X. Xie, and Z. Li, “Drn: A deep reinforcement learning framework for news recommendation,” in Proceedings of the 2018 world wide web conference, 2018, pp. 167–176

    2018

  4. [3]

    Personalized news recommendation: Methods and challenges,

    C. Wu, F. Wu, Y . Huang, and X. Xie, “Personalized news recommendation: Methods and challenges,” ACM Transactions on Information Systems , vol. 41, no. 1, pp. 1–50, 2023

    2023

  5. [4]

    DKN: Deep knowledge-aware network for news recommendation,

    H. Wang, F. Zhang, X. Xie, and M. Guo, “DKN: Deep knowledge-aware network for news recommendation,” inWWW, 2018, pp. 1835–1844

    2018

  6. [5]

    Fine-grained interest matching for neural news recommendation,

    H. Wang, F. Wu, Z. Liu, and X. Xie, “Fine-grained interest matching for neural news recommendation,” in ACL, 2020, pp. 836–845

    2020

  7. [6]

    Neural news recommendation with topic-aware news representation,

    C. Wu, F. Wu, M. An, Y . Huang, and X. Xie, “Neural news recommendation with topic-aware news representation,” inACL, 2019, pp. 1154–1159

    2019

  8. [7]

    Neural news recommendation with long-and short-term user represen- tations,

    M. An, F. Wu, C. Wu, K. Zhang, Z. Liu, and X. Xie, “Neural news recommendation with long-and short-term user represen- tations,” in ACL, 2019, pp. 336–345

    2019

Show all 104 references
  1. [8]

    News recommendation via multi-interest news sequence modelling,

    R. Wang, S. Wang, W. Lu, and X. Peng, “News recommendation via multi-interest news sequence modelling,” in ICASSP, 2022, pp. 7942–7946

    2022

  2. [9]

    Graph neural news recommendation with long-term and short-term interest modeling,

    L. Hu, C. Li, C. Shi, C. Yang, and C. Shao, “Graph neural news recommendation with long-term and short-term interest modeling,” Information Processing & Management , vol. 57, no. 2, p. 102142, 2020

    2020

  3. [10]

    Neural news recommen- dation with collaborative news encoding and structural user encoding,

    Z. Mao, X. Zeng, and K.-F. Wong, “Neural news recommen- dation with collaborative news encoding and structural user encoding,” in EMNLP, 2021, pp. 46–55

    2021

  4. [11]

    Graph neural news recommendation with user existing and potential interest modeling,

    Z. Qiu, Y . Hu, and X. Wu, “Graph neural news recommendation with user existing and potential interest modeling,” ACM Trans- actions on Knowledge Discovery from Data , vol. 16, no. 5, pp. 1–17, 2022

    2022

  5. [12]

    Intention-aware user modeling for personalized news recommendation,

    R. Wang, S. Wang, W. Lu, X. Peng, W. Zhang, C. Zheng, and X. Qiao, “Intention-aware user modeling for personalized news recommendation,” in International Conference on Database Systems for Advanced Applications . Springer, 2023, pp. 179– 194

    2023

  6. [13]

    Attention is all you need,

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, . Kaiser, and I. Polosukhin, “Attention is all you need,” in NeurIPS, 2017, pp. 6000–6010

    2017

  7. [14]

    Fastformer: Additive attention can be all you need,

    C. Wu, F. Wu, T. Qi, Y . Huang, and X. Xie, “Fastformer: Additive attention can be all you need,” arXiv preprint arXiv:2108.09084, 2021

    2021 arXiv

  8. [15]

    Empowering news recommendation with pre-trained language models,

    C. Wu, F. Wu, T. Qi, and Y . Huang, “Empowering news recommendation with pre-trained language models,” in SIGIR, 2021, pp. 1652–1656

    2021

  9. [16]

    Multimodal recommender systems: A survey,

    Q. Liu, J. Hu, Y . Xiao, X. Zhao, J. Gao, W. Wang, Q. Li, and J. Tang, “Multimodal recommender systems: A survey,” ACM Comput. Surv., vol. 57, no. 2, Oct. 2024. [Online]. Available: https://doi.org/10.1145/3695461

    2024 doi

  10. [17]

    Bert: Pre-training of deep bidirectional transformers for language understanding,

    J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “Bert: Pre-training of deep bidirectional transformers for language understanding,” in Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologi...

    2019

  11. [18]

    Transformers: State-of-the-art natural language processing,

    T. Wolf, L. Debut, V . Sanh, J. Chaumond, C. Delangue, A. Moi, P. Cistac, T. Rault, R. Louf, M. Funtowicz et al., “Transformers: State-of-the-art natural language processing,” in Proceedings of the 2020 conference on empirical methods in natural language processing: system dem...

    2020

  12. [19]

    A survey on large language models for recommendation,

    L. Wu, Z. Zheng, Z. Qiu, H. Wang, H. Gu, T. Shen, C. Qin, C. Zhu, H. Zhu, Q. Liu et al. , “A survey on large language models for recommendation,” World Wide Web, vol. 27, no. 5, p. 60, 2024

    2024

  13. [20]

    Roberta: A robustly optimized bert pretraining approach,

    Y . Liu, M. Ott, N. Goyal, J. Du, M. Joshi, D. Chen, O. Levy, M. Lewis, L. Zettlemoyer, and V . Stoyanov, “Roberta: A robustly optimized bert pretraining approach,” arXiv preprint arXiv:1907.11692, 2019

    1907 arXiv

  14. [21]

    Bart: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehen- sion,

    M. Lewis, “Bart: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehen- sion,” arXiv preprint arXiv:1910.13461 , 2019

    1910 arXiv

  15. [22]

    Mtrec: Multi-task learning over bert for news recommendation,

    Q. Bi, J. Li, L. Shang, X. Jiang, Q. Liu, and H. Yang, “Mtrec: Multi-task learning over bert for news recommendation,” in Findings of the Association for Computational Linguistics: ACL 2022, 2022, pp. 2663–2669

    2022

  16. [23]

    MINER: Multi-interest matching network for news recommendation,

    J. Li, J. Zhu, Q. Bi, G. Cai, L. Shang, Z. Dong, X. Jiang, and Q. Liu, “MINER: Multi-interest matching network for news recommendation,” in ACL, 2022, pp. 343–352

    2022

  17. [24]

    Unitrec: A unified text-to-text transformer and joint contrastive learning framework for text-based recommendation,

    Z. Mao, H. Wang, Y . Du, and K.-F. Wong, “Unitrec: A unified text-to-text transformer and joint contrastive learning framework for text-based recommendation,” in Proceedings of the 61st An- nual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers),...

    2023

  18. [25]

    Amm: Attentive multi-field matching for news recommendation,

    Q. Zhang, Q. Jia, C. Wang, J. Li, Z. Wang, and X. He, “Amm: Attentive multi-field matching for news recommendation,” in Proceedings of the 44th international ACM SIGIR conference on research and development in information retrieval, 2021, pp. 1588–1592

    2021

  19. [26]

    News- bert: Distilling pre-trained language model for intelligent news application,

    C. Wu, F. Wu, Y . Yu, T. Qi, Y . Huang, and Q. Liu, “News- bert: Distilling pre-trained language model for intelligent news application,” arXiv preprint arXiv:2102.04887 , 2021

    2021 arXiv

  20. [27]

    Unbert: User-news matching bert for news recommendation

    Q. Zhang, J. Li, Q. Jia, C. Wang, J. Zhu, Z. Wang, and X. He, “Unbert: User-news matching bert for news recommendation.” in IJCAI, vol. 21, 2021, pp. 3356–3362

    2021

  21. [28]

    Generative news recommendation,

    S. Gao, J. Fang, Q. Tu, Z. Yao, Z. Chen, P. Ren, and Z. Ren, “Generative news recommendation,” in Proceedings of the ACM on Web Conference 2024 , 2024, pp. 3444–3453

    2024

  22. [29]

    Rewriting bias: Mitigating media bias in news recommender systems through automated rewriting,

    Q. Ruan, J. Xu, S. Leavy, B. Mac Namee, and R. Dong, “Rewriting bias: Mitigating media bias in news recommender systems through automated rewriting,” in Proceedings of the 32nd ACM Conference on User Modeling, Adaptation and Personalization, 2024, pp. 67–77

    2024

  23. [30]

    Lkpnr: Large language models and knowledge graph for personalized news recommendation framework

    H. Chen, R. Xie, X. Cui, Z. Yan, X. Wang, Z. Xuan, and K. Zhang, “Lkpnr: Large language models and knowledge graph for personalized news recommendation framework.”Computers, Materials & Continua , vol. 79, no. 3, 2024

    2024

  24. [31]

    A survey of personalized news recommendation,

    X. Meng, H. Huo, X. Zhang, W. Wang, and J. Zhu, “A survey of personalized news recommendation,” Data Science and Engineering, vol. 8, no. 4, pp. 396–416, 2023

    2023

  25. [32]

    A survey on knowledge- aware news recommender systems,

    A. Iana, M. Alam, and H. Paulheim, “A survey on knowledge- aware news recommender systems,” Semantic Web, no. Preprint, pp. 1–62, 2024

    2024

  26. [33]

    Dan: Deep atten- tion neural network for news recommendation,

    Q. Zhu, X. Zhou, Z. Song, J. Tan, and L. Guo, “Dan: Deep atten- tion neural network for news recommendation,” in Proceedings of the AAAI conference on artificial intelligence, vol. 33, no. 01, 2019, pp. 5973–5980

    2019

  27. [34]

    Wg4rec: Modeling textual content with word graph for news recommendation,

    S. Shi, W. Ma, Z. Wang, M. Zhang, K. Fang, J. Xu, Y . Liu, and S. Ma, “Wg4rec: Modeling textual content with word graph for news recommendation,” in Proceedings of the 30th ACM International Conference on Information & Knowledge Management, 2021, pp. 1651–1660

    2021

  28. [35]

    Hram: A hybrid recurrent attention machine for news recommendation,

    D. Khattar, V . Kumar, V . Varma, and M. Gupta, “Hram: A hybrid recurrent attention machine for news recommendation,” in Proceedings of the 27th ACM international conference on information and knowledge management, 2018, pp. 1619–1622

    2018

  29. [36]

    Trnews: Heterogeneous user-interest transfer learning for news recommendation,

    G. Hu and Q. Yang, “Trnews: Heterogeneous user-interest transfer learning for news recommendation,” in Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume , 2021, pp. 734– 744

    2021

  30. [37]

    Embedding-based news recommendation for millions of users,

    S. Okura, Y . Tagami, S. Ono, and A. Tajima, “Embedding-based news recommendation for millions of users,” in KDD, 2017, pp. 1933–1942

    2017

  31. [38]

    Time-aware attentive neural network for news recommendation with long-and short- term user representation,

    Y . Pang, Y . Zhang, J. Tong, and Z. Wei, “Time-aware attentive neural network for news recommendation with long-and short- term user representation,” in Knowledge Science, Engineering and Management: 13th International Conference, KSEM 2020, Hangzhou, China, August 28–30, 202...

    2020

  32. [39]

    News recommendation with topic-enriched knowledge graphs,

    D. Lee, B. Oh, S. Seo, and K.-H. Lee, “News recommendation with topic-enriched knowledge graphs,” in Proceedings of the 29th ACM international conference on information & knowledge management, 2020, pp. 695–704

    2020

  33. [40]

    Weave&rec: A word embedding based 3-d convolutional network for news recommendation,

    D. Khattar, V . Kumar, V . Varma, and M. Gupta, “Weave&rec: A word embedding based 3-d convolutional network for news recommendation,” in Proceedings of the 27th ACM international conference on information and knowledge management , 2018, pp. 1855–1858

    2018

  34. [41]

    Deep neural networks for news recommendations,

    K. Park, J. Lee, and J. Choi, “Deep neural networks for news recommendations,” in Proceedings of the 2017 ACM on Conference on Information and Knowledge Management , 2017, pp. 2255–2258

    2017

  35. [42]

    Chameleon: a deep learning meta-architecture for news recommender systems,

    G. de Souza Pereira Moreira, “Chameleon: a deep learning meta-architecture for news recommender systems,” in Proceed- ings of the 12th ACM Conference on Recommender Systems , 2018, pp. 578–583

    2018

  36. [43]

    Neural news recommendation with multi-head self-attention,

    C. Wu, F. Wu, S. Ge, T. Qi, Y . Huang, and X. Xie, “Neural news recommendation with multi-head self-attention,” in EMNLP, 2019, pp. 6390–6395

    2019

  37. [44]

    Dynamic attention-integrated neural network for session-based news recommendation,

    L. Zhang, P. Liu, and J. A. Gulla, “Dynamic attention-integrated neural network for session-based news recommendation,” Ma- chine Learning, vol. 108, pp. 1851–1875, 2019

    2019

  38. [45]

    Dynamic news recommenda- tion with hierarchical attention network,

    H. Zhang, X. Chen, and S. Ma, “Dynamic news recommenda- tion with hierarchical attention network,” in 2019 IEEE Inter- national Conference on Data Mining (ICDM) . IEEE, 2019, pp. 1456–1461

    2019

  39. [46]

    Si-news: Integrating social information for news recommendation with attention-based graph convolutional net- work,

    P. Zhu, D. Cheng, S. Luo, F. Yang, Y . Luo, W. Qian, and A. Zhou, “Si-news: Integrating social information for news recommendation with attention-based graph convolutional net- work,” Neurocomputing, vol. 494, pp. 33–42, 2022

    2022

  40. [47]

    Graph enhanced representation learning for news recommendation,

    S. Ge, C. Wu, F. Wu, T. Qi, and Y . Huang, “Graph enhanced representation learning for news recommendation,” in Proceed- ings of the web conference 2020 , 2020, pp. 2863–2869

    2020

  41. [48]

    Joint knowledge pruning and recurrent graph convolution for news recommendation,

    Y . Tian, Y . Yang, X. Ren, P. Wang, F. Wu, Q. Wang, and C. Li, “Joint knowledge pruning and recurrent graph convolution for news recommendation,” in Proceedings of the 44th international ACM SIGIR conference on research and development in infor- mation retrieval, 2021, pp. 51–60

    2021

  42. [49]

    Rein- forced anchor knowledge graph generation for news recommen- dation reasoning,

    D. Liu, J. Lian, Z. Liu, X. Wang, G. Sun, and X. Xie, “Rein- forced anchor knowledge graph generation for news recommen- dation reasoning,” in Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining , 2021, pp. 1055–1065

    2021

  43. [50]

    Kred: Knowledge-aware document representation for news recommendations,

    D. Liu, J. Lian, S. Wang, Y . Qiao, J.-H. Chen, G. Sun, and X. Xie, “Kred: Knowledge-aware document representation for news recommendations,” in Proceedings of the 14th ACM conference on recommender systems , 2020, pp. 200–209

    2020

  44. [51]

    Knowledge-guided article embedding refinement for session-based news recommenda- tion,

    H.-S. Sheu, Z. Chu, D. Qi, and S. Li, “Knowledge-guided article embedding refinement for session-based news recommenda- tion,” IEEE Transactions on Neural Networks and Learning Systems, vol. 33, no. 12, pp. 7921–7927, 2021

    2021

  45. [52]

    Neural news recommendation with attentive multi-view learning,

    C. Wu, F. Wu, M. An, J. Huang, and Y . Huang, “Neural news recommendation with attentive multi-view learning,” in IJCAI, 2019, pp. 3863–3869

    2019

  46. [53]

    Aspect-driven user preference and news representation learning for news recommendation,

    W. Lu, R. Wang, S. Wang, X. Peng, H. Wu, and Q. Zhang, “Aspect-driven user preference and news representation learning for news recommendation,” IEEE Transactions on Intelligent Transportation Systems , vol. 23, no. 12, pp. 25 297–25 307, 2022

    2022

  47. [54]

    Hi-fi ark: Deep user representation via high-fidelity archive network

    Z. Liu, Y . Xing, F. Wu, M. An, and X. Xie, “Hi-fi ark: Deep user representation via high-fidelity archive network.” in IJCAI, 2019, pp. 3059–3065

    2019

  48. [55]

    Neural news recommendation with heterogeneous user behavior,

    C. Wu, F. Wu, M. An, T. Qi, J. Huang, Y . Huang, and X. Xie, “Neural news recommendation with heterogeneous user behavior,” in Proceedings of the 2019 conference on empirical methods in natural language processing and the 9th international joint conference on natural language ...

    2019

  49. [56]

    Towards better rep- resentation learning for personalized news recommendation: A multi-channel deep fusion approach,

    J. Lian, F. Zhang, X. Xie, and G. Sun, “Towards better rep- resentation learning for personalized news recommendation: A multi-channel deep fusion approach,” in IJCAI, 2018, pp. 3805– 3811

    2018

  50. [57]

    HieRec: Hierarchical user interest modeling for personalized news recommendation,

    T. Qi, F. Wu, C. Wu, P. Yang, Y . Yu, X. Xie, and Y . Huang, “HieRec: Hierarchical user interest modeling for personalized news recommendation,” in ACL, 2021, pp. 5446–5456

    2021

  51. [58]

    Pp-rec: News rec- ommendation with personalized user interest and time-aware news popularity,

    T. Qi, F. Wu, C. Wu, and Y . Huang, “Pp-rec: News rec- ommendation with personalized user interest and time-aware news popularity,” in Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural...

    2021

  52. [59]

    News recommendation with candidate-aware user mod- eling,

    ——, “News recommendation with candidate-aware user mod- eling,” in SIGIR, 2022, p. 19171921

    2022

  53. [60]

    Popularity-enhanced news recommendation with multi-view interest representation,

    J. Wang, Y . Chen, Z. Wang, and W. Zhao, “Popularity-enhanced news recommendation with multi-view interest representation,” in CIKM, 2021, pp. 1949–1958

    2021

  54. [61]

    Intention nets: Psychology-inspired user choice behavior mod- eling for next-basket prediction,

    S. Wang, L. Hu, Y . Wang, Q. Z. Sheng, M. Orgun, and L. Cao, “Intention nets: Psychology-inspired user choice behavior mod- eling for next-basket prediction,” inAAAI, 2020, pp. 6259–6266

    2020

  55. [62]

    Sentirec: Sentiment diversity-aware neural news recommendation,

    C. Wu, F. Wu, T. Qi, and Y . Huang, “Sentirec: Sentiment diversity-aware neural news recommendation,” in Proceedings of the 1st conference of the Asia-Pacific chapter of the associ- ation for computational linguistics and the 10th international joint conference on natural lang...

    2020

  56. [63]

    Removing ais sentiment manipulation of personalized news delivery,

    C. Wu, F. Wu, T. Qi, W.-Q. Zhang, X. Xie, and Y . Huang, “Removing ais sentiment manipulation of personalized news delivery,” Humanities and Social Sciences Communications , vol. 9, no. 1, pp. 1–9, 2022

    2022

  57. [64]

    Fairness-aware news recommendation with decomposed adversarial learning,

    C. Wu, F. Wu, X. Wang, Y . Huang, and X. Xie, “Fairness-aware news recommendation with decomposed adversarial learning,” in Proceedings of the AAAI conference on artificial intelligence, vol. 35, no. 5, 2021, pp. 4462–4469

    2021

  58. [65]

    A hierarchical and disentangling interest learning framework for unbiased and true news recommendation,

    S. Wang, W. Wang, X. Zhang, Y . Wang, H. Liu, and F. Chen, “A hierarchical and disentangling interest learning framework for unbiased and true news recommendation,” in Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2024, pp. 3200–3211

    2024

  59. [66]

    Privacy- preserving news recommendation model learning,

    T. Qi, F. Wu, C. Wu, Y . Huang, and X. Xie, “Privacy- preserving news recommendation model learning,” arXiv preprint arXiv:2003.09592, 2020

    2003 arXiv

  60. [67]

    Efficient- fedrec: Efficient federated learning framework for privacy- preserving news recommendation,

    J. Yi, F. Wu, C. Wu, R. Liu, G. Sun, and X. Xie, “Efficient- fedrec: Efficient federated learning framework for privacy- preserving news recommendation,” in Proceedings of the 2021 Conference on Empirical Methods in Natural Language Pro- cessing, 2021, pp. 2814–2824

    2021

  61. [68]

    User modeling with click preference and reading satisfaction for news recommendation

    C. Wu, F. Wu, T. Qi, and Y . Huang, “User modeling with click preference and reading satisfaction for news recommendation.” in IJCAI, 2020, pp. 3023–3029

    2020

  62. [69]

    Personal or general? a hybrid strategy with multi-factors for news recommendation,

    Z. Huang, B. Jin, H. Zhao, Q. Liu, D. Lian, B. Tengfei, and E. Chen, “Personal or general? a hybrid strategy with multi-factors for news recommendation,” ACM Transactions on Information Systems, vol. 41, no. 2, pp. 1–29, 2023

    2023

  63. [70]

    Train once, use flexibly: A modular framework for multi-aspect neural news recommen- dation,

    A. Iana, G. Glava ˇs, and H. Paulheim, “Train once, use flexibly: A modular framework for multi-aspect neural news recommen- dation,” arXiv preprint arXiv:2307.16089 , 2023

    2023 arXiv

  64. [72]

    News recommendation with cate- gory description by a large language model,

    Y . Yada and H. Yamana, “News recommendation with cate- gory description by a large language model,” arXiv preprint arXiv:2405.13007, 2024

    2024 arXiv

  65. [73]

    Once: Boost- ing content-based recommendation with both open-and closed- source large language models,

    Q. Liu, N. Chen, T. Sakai, and X.-M. Wu, “Once: Boost- ing content-based recommendation with both open-and closed- source large language models,” in Proceedings of the 17th ACM International Conference on Web Search and Data Mining , 2024, pp. 452–461

    2024

  66. [74]

    Modeling user viewing flow using large language models for article recommendation,

    Z. Liu, Z. Chen, M. Zhang, S. Duan, H. Wen, L. Li, N. Li, Y . Gu, and G. Yu, “Modeling user viewing flow using large language models for article recommendation,” in Companion Proceedings of the ACM on Web Conference 2024 , 2024, pp. 83–92. JOURNAL OF LATEX CLASS FILES, VOL. 14...

    2024

  67. [75]

    Tiny-newsrec: Effective and efficient plm-based news recommendation,

    Y . Yu, F. Wu, C. Wu, J. Yi, and Q. Liu, “Tiny-newsrec: Effective and efficient plm-based news recommendation,” in Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, 2022, pp. 5478–5489

    2022

  68. [76]

    Training large-scale news recommenders with pre- trained language models in the loop,

    S. Xiao, Z. Liu, Y . Shao, T. Di, B. Middha, F. Wu, and X. Xie, “Training large-scale news recommenders with pre- trained language models in the loop,” in Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2022, pp. 4215–4225

    2022

  69. [77]

    Recprompt: A prompt tuning framework for news recommendation using large language models,

    D. Liu, B. Yang, H. Du, D. Greene, A. Lawlor, R. Dong, and I. Li, “Recprompt: A prompt tuning framework for news recommendation using large language models,” arXiv preprint arXiv:2312.10463, 2023

    2023 arXiv

  70. [78]

    Cherryrec: Enhancing news recommendation quality via llm-driven framework,

    S. Wang, L. Wang, Y . Bu, and T. Huang, “Cherryrec: Enhancing news recommendation quality via llm-driven framework,” arXiv preprint arXiv:2406.12243, 2024

    2024 arXiv

  71. [79]

    Streamingrec: a framework for benchmarking stream-based news recom- menders,

    M. Jugovac, D. Jannach, and M. Karimi, “Streamingrec: a framework for benchmarking stream-based news recom- menders,” in Proceedings of the 12th ACM conference on recommender systems, 2018, pp. 269–273

    2018

  72. [80]

    Accurate news recommen- dation coalescing personal and global temporal preferences,

    B. Koo, H. Jeon, and U. Kang, “Accurate news recommen- dation coalescing personal and global temporal preferences,” in Advances in Knowledge Discovery and Data Mining: 24th Pacific-Asia Conference, PAKDD 2020, Singapore, May 11–14, 2020, Proceedings, Part I 24 . Springer, 2020...

    2020

  73. [81]

    Dynamic hierarchical attention network for news recommendation,

    Q. Zhao, X. Chen, H. Zhang, and X. Li, “Dynamic hierarchical attention network for news recommendation,” Expert Systems with Applications, vol. 255, p. 124667, 2024

    2024

  74. [82]

    Fair multi- stakeholder news recommender system with hypergraph rank- ing,

    A. Gharahighehi, C. Vens, and K. Pliakos, “Fair multi- stakeholder news recommender system with hypergraph rank- ing,” Information Processing & Management , vol. 58, no. 5, p. 102663, 2021

    2021

  75. [83]

    Lancer: A lifetime- aware news recommender system,

    H.-K. Bae, J. Ahn, D. Lee, and S.-W. Kim, “Lancer: A lifetime- aware news recommender system,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 37, no. 4, 2023, pp. 4141–4148

    2023

  76. [84]

    News session-based recommendations using deep neural net- works,

    G. de Souza Pereira Moreira, F. Ferreira, and A. M. Da Cunha, “News session-based recommendations using deep neural net- works,” in Proceedings of the 3rd workshop on deep learning for recommender systems , 2018, pp. 15–23

    2018

  77. [85]

    The plista dataset,

    B. Kille, F. Hopfgartner, T. Brodt, and T. Heintz, “The plista dataset,” in Proceedings of the 2013 international news recom- mender systems workshop and challenge , 2013, pp. 16–23

    2013

  78. [86]

    The adressa dataset for news recommendation,

    J. A. Gulla, L. Zhang, P. Liu, ¨O. ¨Ozg¨obek, and X. Su, “The adressa dataset for news recommendation,” in Proceedings of the international conference on web intelligence , 2017, pp. 1042–1048

    2017

  79. [87]

    MIND: A large-scale dataset for news recommendation,

    F. Wu, Y . Qiao, J.-H. Chen, T. Wu, Chuhan Qi, J. Lian, D. Liu, X. Xie, J. Gao, and W. Wu, “MIND: A large-scale dataset for news recommendation,” in ACL, 2020, pp. 3597–3606

    2020

  80. [88]

    Npr: a news portal recommendations dataset

    J. P. Lucas, J. F. G. da Silva, and L. F. de Figueiredo, “Npr: a news portal recommendations dataset.” inNORMalize@ RecSys, 2023

    2023

  81. [89]

    Mind your language: A multilingual dataset for cross-lingual news recommendation,

    A. Iana, G. Glava ˇs, and H. Paulheim, “Mind your language: A multilingual dataset for cross-lingual news recommendation,” in Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval , 2024, pp. 553–563

    2024

  82. [90]

    Eb-nerd a large-scale dataset for news recommendation,

    J. Kruse, K. Lindskow, S. Kalloori, M. Polignano, C. Pomo, A. Srivastava, A. Uppal, M. R. Andersen, and J. Frellsen, “Eb-nerd a large-scale dataset for news recommendation,” in Proceedings of the Recommender Systems Challenge 2024 , 2024, pp. 1–11

    2024

  83. [91]

    Mm-rec: Visiolinguistic model empowered multimodal news recommendation,

    C. Wu, F. Wu, T. Qi, C. Zhang, Y . Huang, and T. Xu, “Mm-rec: Visiolinguistic model empowered multimodal news recommendation,” in Proceedings of the 45th international ACM SIGIR conference on research and development in information retrieval, 2022, pp. 2560–2564

    2022

  84. [92]

    Legommenders: A comprehen- sive content-based recommendation library with llm support,

    Q. Liu, L. Fan, and X.-M. Wu, “Legommenders: A comprehen- sive content-based recommendation library with llm support,” arXiv preprint arXiv:2412.15973 , 2024

    2024 arXiv

  85. [93]

    Microsoft recommenders: tools to accelerate developing recommender systems,

    S. Graham, J.-K. Min, and T. Wu, “Microsoft recommenders: tools to accelerate developing recommender systems,” in Pro- ceedings of the 13th ACM Conference on Recommender Sys- tems, 2019, pp. 542–543

    2019

  86. [94]

    Recbole: Towards a unified, comprehensive and efficient framework for recommendation algorithms,

    W. X. Zhao, S. Mu, Y . Hou, Z. Lin, Y . Chen, X. Pan, K. Li, Y . Lu, H. Wang, C. Tian, Y . Min, Z. Feng, X. Fan, X. Chen, P. Wang, W. Ji, Y . Li, X. Wang, and J. Wen, “Recbole: Towards a unified, comprehensive and efficient framework for recommendation algorithms,” in CIKM. AC...

    2021

  87. [95]

    Spar: Personalized content-based recommendation via long engagement attention,

    C. Zhang, Y . Sun, J. Chen, J. Lei, M. Abdul-Mageed, S. Wang, R. Jin, S. Park, N. Yao, and B. Long, “Spar: Personalized content-based recommendation via long engagement attention,” arXiv preprint arXiv:2402.10555 , 2024

    2024 arXiv

  88. [96]

    Benchmarking news recommendation in the era of green ai,

    Q. Liu, J. Zhu, Q. Dai, and X.-M. Wu, “Benchmarking news recommendation in the era of green ai,” in Companion Proceed- ings of the ACM on Web Conference 2024 , 2024, pp. 971–974

    2024

  89. [97]

    Discrete semantic tokenization for deep ctr prediction,

    Q. Liu, H. Hu, J. Wu, J. Zhu, M.-Y . Kan, and X.-M. Wu, “Discrete semantic tokenization for deep ctr prediction,” in Companion Proceedings of the ACM on Web Conference 2024 , 2024, pp. 919–922

    2024

  90. [98]

    Halogen: Fantastic llm hallucinations and where to find them,

    A. Ravichander, S. Ghela, D. Wadden, and Y . Choi, “Halogen: Fantastic llm hallucinations and where to find them,” arXiv preprint arXiv:2501.08292, 2025

    2025 arXiv

  91. [99]

    Enhanc- ing llm’s ability to generate more repository-aware unit tests through precise contextual information injection,

    X. Yin, C. Ni, X. Li, L. Chen, G. Ma, and X. Yang, “Enhanc- ing llm’s ability to generate more repository-aware unit tests through precise contextual information injection,”arXiv preprint arXiv:2501.07425, 2025

    2025 arXiv

  92. [100]

    Prompt learning for news recom- mendation,

    Z. Zhang and B. Wang, “Prompt learning for news recom- mendation,” in Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2023, pp. 227–237

    2023

  93. [101]

    Prompt-based generative news recommendation (pgnr): Accuracy and controllability,

    X. Li, Y . Zhang, and E. C. Malthouse, “Prompt-based generative news recommendation (pgnr): Accuracy and controllability,” in European Conference on Information Retrieval. Springer, 2024, pp. 66–79

    2024

  94. [102]

    Multi-agent collaboration framework for recommender systems,

    Z. Wang, Y . Yu, W. Zheng, W. Ma, and M. Zhang, “Multi-agent collaboration framework for recommender systems,” arXiv preprint arXiv:2402.15235, 2024

    2024 arXiv

  95. [103]

    When large lan- guage model based agent meets user behavior analysis: A novel user simulation paradigm,

    L. Wang, J. Zhang, H. Yang, Z. Chen, J. Tang, Z. Zhang, X. Chen, Y . Lin, R. Song, W. X. Zhao et al., “When large lan- guage model based agent meets user behavior analysis: A novel user simulation paradigm,” arXiv preprint arXiv:2306.02552 , 2023

    2023 arXiv

  96. [104]

    An efficient recommen- dation generation using relevant jaccard similarity,

    S. Bag, S. K. Kumar, and M. K. Tiwari, “An efficient recommen- dation generation using relevant jaccard similarity,” Information Sciences, vol. 483, pp. 53–64, 2019

    2019

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.