pith. sign in

arxiv: 2606.00540 · v1 · pith:LJL6ZJZUnew · submitted 2026-05-30 · 💻 cs.IR

Trustworthy Recommendation in the Era of Large Language Models: Opportunities and Challenges

Bohao Wang , Yu Cui , Zhenxiang Xu , Jujia Zhao , Chenxiao Fan , Jizhi Zhang , Weiqin Yang , Shengjia Zhang
show 8 more authors
Sirui Chen Yang Zhang Xiaoyan Zhao Wenjie Wang Chongming Gao Fuli Feng Xiangnan He Jiawei Chen
This is my paper

Pith reviewed 2026-06-28 18:39 UTC · model grok-4.3

classification 💻 cs.IR
keywords recommender systemslarge language modelstrustworthinessfairnessrobustnessprivacybiashallucinations
0
0 comments X

The pith

Large language models act as a double-edged sword for trustworthy recommendation systems by enabling new capabilities while creating fresh risks.

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

Recommender systems have shifted focus from raw accuracy toward trustworthiness dimensions such as robustness, fairness, and privacy. Large language models are now widely integrated into these systems, supplying richer semantic understanding and flexible interactions. A review of more than 200 studies concludes that this integration produces both gains and drawbacks: LLMs can strengthen trustworthiness through advanced reasoning yet also inject novel biases and hallucination problems. The work organizes the literature into a taxonomy of 13 opportunities and 18 challenges spread across six core trustworthiness dimensions. It further surveys datasets, metrics, and remaining open questions to guide empirical work.

Core claim

The integration of large language models into recommender systems produces a dual effect on trustworthiness. While the models supply mechanisms for stronger intent reasoning and user interaction that can improve robustness, fairness, and privacy, they simultaneously introduce new failure modes including amplified bias and hallucination-driven errors. Systematic analysis of over 200 studies yields a taxonomy that identifies 13 opportunities and 18 challenges across six fundamental dimensions, together with a survey of supporting datasets and evaluation metrics.

What carries the argument

A taxonomy classifying existing literature on LLM-empowered recommendation into 13 opportunities and 18 challenges distributed across six trustworthiness dimensions.

If this is right

  • Evaluation protocols must incorporate new metrics that detect LLM-specific issues such as hallucination in generated recommendations.
  • Mitigation strategies are required to address novel bias forms introduced by language-model reasoning.
  • Privacy techniques need adaptation to handle the semantic inference capabilities of LLMs.
  • Fairness definitions should be revised to account for biases that arise from LLM training data and generation processes.
  • Future systems will need hybrid designs that selectively apply LLM components only where their benefits exceed the added risks.

Where Pith is reading between the lines

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

  • The taxonomy may need extension once multimodal large language models become common in recommendation pipelines.
  • High-stakes applications such as medical or financial recommendations may require extra verification layers not yet captured in the current challenge list.
  • Transferable techniques from general large-model safety research could address some of the hallucination and bias challenges identified here.
  • Large-scale experiments on the reviewed datasets could quantify whether the net effect of LLMs on trustworthiness is positive or negative in practice.

Load-bearing premise

A comprehensive review of over 200 studies produces an unbiased and gap-free taxonomy of all relevant opportunities and challenges.

What would settle it

Discovery of a sizable set of studies on LLM-based recommendation whose findings fall outside the proposed 13 opportunities and 18 challenges or that reveal major uncovered areas in the six dimensions.

Figures

Figures reproduced from arXiv: 2606.00540 by Bohao Wang, Chenxiao Fan, Chongming Gao, Fuli Feng, Jiawei Chen, Jizhi Zhang, Jujia Zhao, Shengjia Zhang, Sirui Chen, Weiqin Yang, Wenjie Wang, Xiangnan He, Xiaoyan Zhao, Yang Zhang, Yu Cui, Zhenxiang Xu.

Figure 1
Figure 1. Figure 1: Overview of this survey structure. recommendation. Therefore, we believe it is both timely and imperative to present a comprehensive survey of LLM￾empowered trustworthy recommendation. Contributions. In this work, we provide a systematic review of this growing research area, focusing primarily on how the integration of LLMs changes the landscape of trustworthy recommendation. By comprehensively analyzing o… view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of the paradigms of traditional recommender systems and LLM-empowered recommenders. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Six key dimensions of trustworthy recommendation in the era of large language models. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Opportunities and challenges of LLM-empowered recommendation across six dimensions of trustworthiness. Red boxes denote [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: An illustration of the opportunities and challenges introduced by LLMs for the robustness of recommender systems. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: An illustration of the opportunities and challenges introduced by LLMs for the bias and fairness of recommender systems. [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: An illustration of the opportunities and challenges introduced by LLMs for the explainability of recommender systems. [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: An illustration of the challenges introduced by LLMs for the factuality of recommender systems. [PITH_FULL_IMAGE:figures/full_fig_p018_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: An illustration of the opportunities and challenges introduced by LLMs for the controllability of recommender systems. [PITH_FULL_IMAGE:figures/full_fig_p020_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: An illustration of the opportunities and challenges introduced by LLMs for the privacy of recommender systems. [PITH_FULL_IMAGE:figures/full_fig_p023_10.png] view at source ↗
read the original abstract

The field of recommender systems (RS) is currently undergoing two profound paradigm shifts. From the perspective of objectives, the goal has shifted beyond mere recommendation accuracy to comprehensive trustworthiness, encompassing multiple dimensions such as robustness, fairness, and privacy preservation. From a technical perspective, Large Language Models (LLMs) have been extensively integrated into RS, reshaping the foundations of recommendation through richer semantic understanding, stronger intent reasoning, and more flexible user interactions. The convergence of these two shifts prompts a timely and pivotal question: how does the integration of LLMs reshape the landscape of trustworthy recommendation? In this work, we present a systematic review of trustworthy LLM-empowered recommendation. By comprehensively analyzing over 200 recent studies, we reveal that the introduction of LLMs acts as a double-edged sword. While their advanced mechanisms and user-friendly interfaces offer unprecedented opportunities to enhance trustworthiness, they simultaneously introduce new risks, such as novel forms of bias and hallucination-induced issues. To characterize this dual impact, we systematically identify 13 opportunities and 18 challenges across six fundamental dimensions of trustworthiness, and accordingly organize the existing literature into a novel taxonomy. We also provide a comprehensive review of commonly used datasets and evaluation metrics to facilitate empirical validation. Finally, we identify critical open challenges and outline future directions, hoping to inspire future research on this emerging topic.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 0 minor

Summary. The paper presents a systematic review of trustworthy LLM-empowered recommender systems. It claims that integrating LLMs into RS creates a double-edged effect: 13 opportunities to improve trustworthiness (across dimensions such as robustness, fairness, and privacy) alongside 18 new challenges (including novel biases and hallucination issues). The authors synthesize findings from over 200 studies into a novel taxonomy, review datasets and evaluation metrics, and outline future research directions.

Significance. A well-executed synthesis of this scope could provide a useful organizing framework for an emerging intersection of LLMs and trustworthy RS, helping researchers navigate both enhancements and risks. The explicit enumeration of opportunities and challenges, together with the dataset/metric review, would be particularly valuable if the underlying corpus selection is reproducible.

major comments (2)
  1. [Abstract] Abstract: The central claim—that a comprehensive analysis of >200 studies yields a complete taxonomy of 13 opportunities and 18 challenges across six trustworthiness dimensions—rests on an undocumented literature review process. No search strategy, databases, keyword strings, date cutoffs, inclusion/exclusion criteria, or inter-rater reliability measures are described, leaving the taxonomy vulnerable to selection bias and undermining reproducibility of the opportunity/challenge mapping.
  2. [Abstract] The headline characterization of LLMs as a 'double-edged sword' is derived entirely from the synthesis; without the missing methodological details, it is impossible to assess whether counter-examples or underrepresented venues were systematically omitted, which directly affects the load-bearing claim that the taxonomy is representative.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for highlighting the need for greater transparency in our literature review process. We agree that explicit methodological details are required to support the reproducibility and representativeness claims in a systematic review. We address both comments below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim—that a comprehensive analysis of >200 studies yields a complete taxonomy of 13 opportunities and 18 challenges across six trustworthiness dimensions—rests on an undocumented literature review process. No search strategy, databases, keyword strings, date cutoffs, inclusion/exclusion criteria, or inter-rater reliability measures are described, leaving the taxonomy vulnerable to selection bias and undermining reproducibility of the opportunity/challenge mapping.

    Authors: We agree that the current manuscript lacks a dedicated description of the literature search and selection process. This omission weakens the ability to evaluate selection bias and reproducibility. In the revision we will add a new 'Review Methodology' subsection (likely in Section 2 or as an appendix) that explicitly documents: (1) the databases and repositories searched (Google Scholar, arXiv, ACM DL, IEEE Xplore), (2) the keyword strings and Boolean combinations used, (3) the time window (primarily 2022 onward for LLM-related work), (4) inclusion/exclusion criteria, (5) the screening and coding procedure, and (6) any inter-rater checks performed. This will allow readers to assess how the 13 opportunities and 18 challenges were mapped from the corpus. revision: yes

  2. Referee: [Abstract] The headline characterization of LLMs as a 'double-edged sword' is derived entirely from the synthesis; without the missing methodological details, it is impossible to assess whether counter-examples or underrepresented venues were systematically omitted, which directly affects the load-bearing claim that the taxonomy is representative.

    Authors: The double-edged-sword framing is an interpretive summary of the opportunities and challenges identified across the reviewed papers rather than an independent empirical claim. Nevertheless, we accept that without documented search and selection procedures it is difficult to judge coverage or the risk of omitted counter-examples. The added methodology section will include a limitations paragraph that discusses venue coverage, the recency bias inherent to LLM literature, and any steps taken to mitigate under-representation. We will also make the full list of reviewed papers available as supplementary material to further support scrutiny of the taxonomy. revision: yes

Circularity Check

0 steps flagged

Survey synthesis draws from external literature; no self-referential derivations or load-bearing self-citations

full rationale

The paper is a literature review that organizes >200 external studies into a taxonomy of 13 opportunities and 18 challenges. No equations, fitted parameters, predictions, or uniqueness theorems are present. The central claim (LLMs as double-edged sword) is framed as emerging from analysis of cited external work rather than reducing to any input by construction or author-overlapping citation chain. Self-citation, if present, is not load-bearing for the taxonomy itself. This meets the default expectation of no significant circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

As a literature survey, the paper introduces no new free parameters, mathematical axioms, or invented entities; it relies on standard practices for systematic reviews.

axioms (1)
  • domain assumption Standard practices for conducting systematic literature reviews are followed when selecting and analyzing the 200 studies.
    The abstract claims a comprehensive analysis without detailing deviations from established review protocols.

pith-pipeline@v0.9.1-grok · 5817 in / 1230 out tokens · 30159 ms · 2026-06-28T18:39:33.195665+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

300 extracted references · 92 canonical work pages · 15 internal anchors

  1. [1]

    Himan Abdollahpouri, Gediminas Adomavicius, Robin Burke, Ido Guy, Dietmar Jannach, Toshihiro Kamishima, Jan Krasnodebski, and Luiz Pizzato

  2. [2]

    Abdollahpouri et al.User Modeling and User-Adapted Interaction30, 1 (2020), 127–158

    Multistakeholder recommendation: Survey and research directions: H. Abdollahpouri et al.User Modeling and User-Adapted Interaction30, 1 (2020), 127–158

    2020

  3. [3]

    Hasan Abu-Rasheed, Christian Weber, and Madjid Fathi. 2024. Knowledge graphs as context sources for llm-based explanations of learning recommendations. In2024 IEEE Global Engineering Education Conference (EDUCON). IEEE, 1–5

    2024

  4. [4]

    Aaron Adcock, Aayushi Srivastava, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pande, Abhinav Pandey, Abhinav Sharma, Abhishek Kadian, Abhishek Kumawat, Adam Kelsey, et al. 2026. The Llama 4 Herd: Architecture, Training, Evaluation, and Deployment Notes.arXiv preprint arXiv:2601.11659(2026)

    work page arXiv 2026

  5. [5]

    Julien Albert, Martin Balfroid, Miriam Doh, Jérémie Bogaert, Luca La Fisca, Liesbet De Vos, Bryan Renard, Vincent Stragier, and Emmanuel Jean. 2024. User Preferences for Large Language Model versus Template-Based Explanations of Movie Recommendations: A Pilot Study.ArXiv abs/2409.06297 (2024). https://api.semanticscholar.org/CorpusID:272550316

    work page arXiv 2024

  6. [6]

    Martin Arjovsky, Léon Bottou, Ishaan Gulrajani, and David Lopez-Paz. 2019. Invariant risk minimization.arXiv preprint arXiv:1907.02893(2019)

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  7. [7]

    Giacomo Balloccu, Ludovico Boratto, Christian Cancedda, Gianni Fenu, and Mirko Marras. 2023. Faithful Path Language Modeling for Explainable Recommendation over Knowledge Graph.arXiv preprint arXiv:2310.16452(2023)

    work page arXiv 2023

  8. [8]

    Yejin Bang, Ziwei Ji, Alan Schelten, Anthony Hartshorn, Tara Fowler, Cheng Zhang, Nicola Cancedda, and Pascale Fung. 2025. Hallulens: Llm hallucination benchmark. InProceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 24128–24156

    2025

  9. [9]

    Keqin Bao, Jizhi Zhang, Wenjie Wang, Yang Zhang, Zhengyi Yang, Yanchen Luo, Chong Chen, Fuli Feng, and Qi Tian. 2025. A bi-step grounding paradigm for large language models in recommendation systems.ACM Transactions on Recommender Systems3, 4 (2025), 1–27

    2025

  10. [10]

    Keqin Bao, Jizhi Zhang, Yang Zhang, Xinyue Huo, Chong Chen, and Fuli Feng. 2024. Decoding matters: Addressing amplification bias and homogeneity issue in recommendations for large language models. InProceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 10540–10552

    2024

  11. [11]

    Keqin Bao, Jizhi Zhang, Yang Zhang, Wenjie Wang, Fuli Feng, and Xiangnan He. 2023. Tallrec: An effective and efficient tuning framework to align large language model with recommendation. InProceedings of the 17th ACM Conference on Recommender Systems. 1007–1014

    2023

  12. [12]

    Milena Belosevic. 2025. User-Centric Design Paradigms for Trust and Control in Human-LLM-Interactions: A Survey. InProceedings of the Fourth Workshop on Bridging Human-Computer Interaction and Natural Language Processing (HCI+ NLP). 17–32

    2025

  13. [13]

    Shai Ben-David, John Blitzer, Koby Crammer, and Fernando Pereira. 2006. Analysis of representations for domain adaptation.Advances in neural information processing systems19 (2006)

    2006

  14. [14]

    Millennium Bismay, Xiangjue Dong, and James Caverlee. 2024. ReasoningRec: Bridging Personalized Recommendations and Human-Interpretable Explanations through LLM Reasoning.arXiv preprint arXiv:2410.23180(2024)

    work page arXiv 2024

  15. [15]

    Ethan Bito, Yongli Ren, and Estrid He. 2025. Evaluating Position Bias in Large Language Model Recommendations.arXiv preprint arXiv:2508.02020 (2025)

    work page arXiv 2025

  16. [16]

    Nicholas Carlini, Daphne Ippolito, Matthew Jagielski, Katherine Lee, Florian Tramer, and Chiyuan Zhang. 2022. Quantifying memorization across neural language models. InThe Eleventh International Conference on Learning Representations

    2022

  17. [17]

    Nicholas Carlini, Florian Tramer, Eric Wallace, Matthew Jagielski, Ariel Herbert-Voss, Katherine Lee, Adam Roberts, Tom Brown, Dawn Song, Ulfar Erlingsson, et al. 2021. Extracting training data from large language models. In30th USENIX security symposium (USENIX Security 21). 2633–2650

    2021

  18. [18]

    Micah Carroll, Adeline Foote, Marcus Williams, Anca Dragan, W Bradley Knox, and Smitha Milli. [n. d.]. CTRL-Rec: Controlling Recommender Systems With Natural Language. InICLR 2025 Workshop on Bidirectional Human-AI Alignment

    2025

  19. [19]

    Yukuo Cen, Jianwei Zhang, Xu Zou, Chang Zhou, Hongxia Yang, and Jie Tang. 2020. Controllable multi-interest framework for recommendation. InProceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining. 2942–2951

    2020

  20. [20]

    Allison JB Chaney, Brandon M Stewart, and Barbara E Engelhardt. 2018. How algorithmic confounding in recommendation systems increases homogeneity and decreases utility. InProceedings of the 12th ACM conference on recommender systems. 224–232

    2018

  21. [21]

    Ting-Jui Chang, Lydia Hsiao-Mei Lin, and Richard Tzong-Han Tsai. 2024. Conversational product recommendation using LLM. In2024 IEEE 4th International Conference on Electronic Communications, Internet of Things and Big Data (ICEIB). IEEE, 340–343. Manuscript submitted to ACM 36 Bohao Wang et al

    2024

  22. [22]

    Shangkun Che, Minjia Mao, and Hongyan Liu. 2024. New Community Cold-Start Recommendation: A Novel Large Language Model-based Method. (2024)

    2024

  23. [23]

    Chong Chen, Fei Sun, Min Zhang, and Bolin Ding. 2022. Recommendation unlearning. InProceedings of the ACM web conference 2022. 2768–2777

    2022

  24. [24]

    Hanxiong Chen, Shaoyun Shi, Yunqi Li, and Yongfeng Zhang. 2021. Neural collaborative reasoning. InProceedings of the web conference 2021. 1516–1527

    2021

  25. [25]

    Junyi Chen, Lu Chi, Bingyue Peng, and Zehuan Yuan. 2024. Hllm: Enhancing sequential recommendations via hierarchical large language models for item and user modeling.arXiv preprint arXiv:2409.12740(2024)

    work page arXiv 2024

  26. [26]

    Jiawei Chen, Hande Dong, Xiang Wang, Fuli Feng, Meng Wang, and Xiangnan He. 2023. Bias and debias in recommender system: A survey and future directions.ACM Transactions on Information Systems41, 3 (2023), 1–39

    2023

  27. [27]

    Jiaju Chen, Chongming Gao, Shuai Yuan, Shuchang Liu, Qingpeng Cai, and Peng Jiang. 2025. DLCRec: A Novel Approach for Managing Diversity in LLM-Based Recommender Systems. InProceedings of the Eighteenth ACM International Conference on Web Search and Data Mining. 857–865

    2025

  28. [28]

    Wei Chen and Dexin Chen. 2025. LLM-as-Critic: Contrastive and Adversarial Strategies for Authentic Text Verification. (2025)

    2025

  29. [29]

    Weixin Chen, Yuhan Zhao, Jingyuan Huang, Zihe Ye, Clark Mingxuan Ju, Tong Zhao, Neil Shah, Li Chen, and Yongfeng Zhang. 2026. MemRec: Collaborative Memory-Augmented Agentic Recommender System.arXiv preprint arXiv:2601.08816(2026)

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  30. [30]

    Hung-Yun Chiang, Yi-Syuan Chen, Yun-Zhu Song, Hong-Han Shuai, and Jason S Chang. 2023. Shilling black-box review-based recommender systems through fake review generation. InProceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 286–297

    2023

  31. [31]

    Chen-Yao Chung, Ping-Yu Hsu, and Shih-Hsiang Huang. 2013. 𝛽P: A novel approach to filter out malicious rating profiles from recommender systems.Decision Support Systems55, 1 (2013), 314–325

    2013

  32. [32]

    Yu Cui, Feng Liu, Pengbo Wang, Bohao Wang, Heng Tang, Yi Wan, Jun Wang, and Jiawei Chen. 2024. Distillation matters: empowering sequential recommenders to match the performance of large language models. InProceedings of the 18th ACM Conference on Recommender Systems. 507–517

    2024

  33. [33]

    Zeyu Cui, Jianxin Ma, Chang Zhou, Jingren Zhou, and Hongxia Yang. 2022. M6-rec: Generative pretrained language models are open-ended recommender systems.arXiv preprint arXiv:2205.08084(2022)

    work page arXiv 2022

  34. [34]

    Ivens da Silva Portugal, Paulo Alencar, and Donald Cowan. 2026. Agentic Recommender Systems: A Systematic Literature Review.IEEE Transactions on Software Engineering(2026)

    2026

  35. [35]

    Sunhao Dai, Chen Xu, Shicheng Xu, Liang Pang, Zhenhua Dong, and Jun Xu. 2024. Bias and unfairness in information retrieval systems: New challenges in the llm era. InProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 6437–6447

    2024

  36. [36]

    Wenxi Dai, Wujiang Xu, Pinhuan Wang, and Dimitris N Metaxas. 2025. Token-Controlled Re-ranking for Sequential Recommendation via LLMs. arXiv preprint arXiv:2511.17913(2025)

    work page arXiv 2025

  37. [37]

    Anindya Bijoy Das and Shahnewaz Karim Sakib. 2024. Unveiling and Mitigating Bias in Large Language Model Recommendations: A Path to Fairness.arXiv preprint arXiv:2409.10825(2024)

    work page arXiv 2024

  38. [38]

    Badhan Chandra Das, M Hadi Amini, and Yanzhao Wu. 2025. Security and privacy challenges of large language models: A survey.Comput. Surveys 57, 6 (2025), 1–39

    2025

  39. [39]

    Yashar Deldjoo. 2024. Understanding biases in ChatGPT-based recommender systems: Provider fairness, temporal stability, and recency.ACM Transactions on Recommender Systems(2024)

    2024

  40. [40]

    Yashar Deldjoo and Tommaso Di Noia. 2025. Cfairllm: Consumer fairness evaluation in large-language model recommender system.ACM Transactions on Intelligent Systems and Technology(2025)

    2025

  41. [41]

    Yashar Deldjoo, Nikhil Mehta, Maheswaran Sathiamoorthy, Shuai Zhang, Pablo Castells, and Julian McAuley. 2025. Toward Holistic Evaluation of Recommender Systems Powered by Generative Models. InProceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval. 3932–3942

    2025

  42. [42]

    Yashar Deldjoo and Fatemeh Nazary. 2024. A Normative Framework for Benchmarking Consumer Fairness in Large Language Model Recommender System.arXiv preprint arXiv:2405.02219(2024)

    work page arXiv 2024

  43. [43]

    Jiaxin Deng, Shiyao Wang, Kuo Cai, Lejian Ren, Qigen Hu, Weifeng Ding, Qiang Luo, and Guorui Zhou. 2025. Onerec: Unifying retrieve and rank with generative recommender and iterative preference alignment.arXiv preprint arXiv:2502.18965(2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  44. [44]

    Dario Di Palma. 2023. Retrieval-augmented recommender system: Enhancing recommender systems with large language models. InProceedings of the 17th ACM Conference on Recommender Systems. 1369–1373

    2023

  45. [45]

    David Dohan, Winnie Xu, Aitor Lewkowycz, Jacob Austin, David Bieber, Raphael Gontijo Lopes, Yuhuai Wu, Henryk Michalewski, Rif A Saurous, Jascha Sohl-Dickstein, et al. 2022. Language model cascades.arXiv preprint arXiv:2207.10342(2022)

    work page arXiv 2022

  46. [46]

    Shen Dong, Shaochen Xu, Pengfei He, Yige Li, Jiliang Tang, Tianming Liu, Hui Liu, and Zhen Xiang. 2025. Memory Injection Attacks on LLM Agents via Query-Only Interaction.arXiv preprint arXiv:2503.03704(2025)

    work page arXiv 2025

  47. [47]

    Yilun Du, Shuang Li, Antonio Torralba, Joshua B Tenenbaum, and Igor Mordatch. 2024. Improving factuality and reasoning in language models through multiagent debate. InForty-first international conference on machine learning

    2024

  48. [48]

    Danial Ebrat, Eli Paradalis, and Luis Rueda. 2024. Lusifer: LLM-based user simulated feedback environment for online recommender systems. arXiv preprint arXiv:2405.13362(2024)

    work page arXiv 2024

  49. [49]

    Qamar EL Maazouzi, Asmaâ Retbi, and Samir Bennani. 2024. Optimizing Recommendation Systems in E-Learning: Synergistic Integration of Lang Chain, GPT Models, and Retrieval Augmented Generation (RAG). InInternational Conference on Smart Applications and Data Analysis. Springer, 106–118. Manuscript submitted to ACM Trustworthy Recommendation in the Era of La...

    2024

  50. [50]

    Chenxiao Fan, Chongming Gao, Yaxin Gong, Haoyan Liu, Fuli Feng, and Xiangnan He. 2026. Uncertainty-aware Generative Recommendation. arXiv preprint arXiv:2602.11719(2026)

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  51. [51]

    Chenxiao Fan, Chongming Gao, Wentao Shi, Yaxin Gong, Zihao Zhao, and Fuli Feng. 2025. Fine-grained List-wise Alignment for Generative Medication Recommendation.arXiv preprint arXiv:2505.20218(2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  52. [52]

    Wenqi Fan, Xiangyu Zhao, Xiao Chen, Jingran Su, Jingtong Gao, Lin Wang, Qidong Liu, Yiqi Wang, Han Xu, Lei Chen, et al. 2022. A comprehensive survey on trustworthy recommender systems.arXiv preprint arXiv:2209.10117(2022)

    work page arXiv 2022

  53. [53]

    Shanshan Feng, Haoming Lyu, Fan Li, Zhu Sun, and Caishun Chen. 2024. Where to move next: Zero-shot generalization of llms for next poi recommendation. In2024 IEEE Conference on Artificial Intelligence (CAI). IEEE, 1530–1535

    2024

  54. [54]

    Miriam Fernandez and Alejandro Bellogin. 2020. Recommender Systems and Misinformation: The Problem or the Solution?. InOHARS Workshop. 14th ACM Conference on Recommender Systems. OHARS Workshop. 14th ACM Conference on Recommender Systems. https://oro.open.ac.uk/72186/ Conference Website: https://recsys.acm.org/ Workshop Website: https://ohars-recsys2020.i...

    2020

  55. [55]

    Giorgos Filandrianos, Angeliki Dimitriou, Maria Lymperaiou, Konstantinos Thomas, and Giorgos Stamou. 2025. Bias Beware: The Impact of Cognitive Biases on LLM-Driven Product Recommendations.arXiv preprint arXiv:2502.01349(2025)

    work page arXiv 2025

  56. [56]

    Luke Friedman, Sameer Ahuja, David Allen, Zhenning Tan, Hakim Sidahmed, Changbo Long, Jun Xie, Gabriel Schubiner, Ajay Patel, Harsh Lara, et al. 2023. Leveraging large language models in conversational recommender systems.arXiv preprint arXiv:2305.07961(2023)

    work page arXiv 2023

  57. [57]

    Isabel O Gallegos, Ryan A Rossi, Joe Barrow, Md Mehrab Tanjim, Sungchul Kim, Franck Dernoncourt, Tong Yu, Ruiyi Zhang, and Nesreen K Ahmed. 2024. Bias and fairness in large language models: A survey.Computational Linguistics50, 3 (2024), 1097–1179

    2024

  58. [58]

    Chongming Gao, Ruijun Chen, Shuai Yuan, Kexin Huang, Yuanqing Yu, and Xiangnan He. 2025. Sprec: Self-play to debias llm-based recommendation. InProceedings of the ACM on Web Conference 2025. 5075–5084

    2025

  59. [59]

    Chongming Gao, Mengyao Gao, Chenxiao Fan, Shuai Yuan, Wentao Shi, and Xiangnan He. 2025. Process-supervised llm recommenders via flow-guided tuning. InProceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1934–1943

    2025

  60. [60]

    Chongming Gao, Wenqiang Lei, Xiangnan He, Maarten De Rijke, and Tat-Seng Chua. 2021. Advances and challenges in conversational recommender systems: A survey.AI open2 (2021), 100–126

    2021

  61. [61]

    Jingtong Gao, Bo Chen, Xiangyu Zhao, Weiwen Liu, Xiangyang Li, Yichao Wang, Zijian Zhang, Wanyu Wang, Yuyang Ye, Shanru Lin, et al. 2024. Llm-enhanced reranking in recommender systems.arXiv preprint arXiv:2406.12433(2024)

    work page arXiv 2024

  62. [62]

    Shen Gao, Yifan Wang, Jiabao Fang, Lisi Chen, Peng Han, and Shuo Shang. 2024. DRE: Generating Recommendation Explanations by Aligning Large Language Models at Data-level.arXiv preprint arXiv:2404.06311(2024)

    work page arXiv 2024

  63. [63]

    Yunfan Gao, Tao Sheng, Youlin Xiang, Yun Xiong, Haofen Wang, and Jiawei Zhang. 2023. Chat-rec: Towards interactive and explainable llms-augmented recommender system.arXiv preprint arXiv:2303.14524(2023)

    work page arXiv 2023

  64. [64]

    Yingqiang Ge, Shuchang Liu, Zuohui Fu, Juntao Tan, Zelong Li, Shuyuan Xu, Yunqi Li, Yikun Xian, and Yongfeng Zhang. 2024. A survey on trustworthy recommender systems.ACM Transactions on Recommender Systems3, 2 (2024), 1–68

    2024

  65. [65]

    Yingqiang Ge, Shuya Zhao, Honglu Zhou, Changhua Pei, Fei Sun, Wenwu Ou, and Yongfeng Zhang. 2020. Understanding echo chambers in e-commerce recommender systems. InProceedings of the 43rd international ACM SIGIR conference on research and development in information retrieval. 2261–2270

    2020

  66. [66]

    Shijie Geng, Zuohui Fu, Juntao Tan, Yingqiang Ge, Gerard De Melo, and Yongfeng Zhang. 2022. Path language modeling over knowledge graphsfor explainable recommendation. InProceedings of the ACM web conference 2022. 946–955

    2022

  67. [67]

    Shijie Geng, Shuchang Liu, Zuohui Fu, Yingqiang Ge, and Yongfeng Zhang. 2022. Recommendation as language processing (rlp): A unified pretrain, personalized prompt & predict paradigm (p5). InProceedings of the 16th ACM conference on recommender systems. 299–315

    2022

  68. [68]

    Preetam Ghosh and Vaishali Sadaphal. 2023. JobRecoGPT–Explainable job recommendations using LLMs.arXiv preprint arXiv:2309.11805(2023)

    work page arXiv 2023

  69. [69]

    Yuqi Gong, Xichen Ding, Yehui Su, Kaiming Shen, Zhongyi Liu, and Guannan Zhang. 2023. An unified search and recommendation foundation model for cold-start scenario. InProceedings of the 32nd ACM International Conference on Information and Knowledge Management. 4595–4601

    2023

  70. [70]

    Yaxin Gong, Chongming Gao, Chenxiao Fan, Wenjie Wang, Fuli Feng, and Xiangnan He. 2026. Breaking User-Centric Agency: A Tri-Party Framework for Agent-Based Recommendation.arXiv preprint arXiv:2603.10673(2026)

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  71. [71]

    Shengkang Gu, Jiahao Liu, Dongsheng Li, Guangping Zhang, Mingzhe Han, Hansu Gu, Peng Zhang, Ning Gu, Li Shang, and Tun Lu. 2025. LLM-Based User Simulation for Low-Knowledge Shilling Attacks on Recommender Systems.arXiv preprint arXiv:2505.13528(2025)

    work page arXiv 2025

  72. [72]

    Ihsan Gunes, Cihan Kaleli, Alper Bilge, and Huseyin Polat. 2014. Shilling attacks against recommender systems: a comprehensive survey.Artificial Intelligence Review42, 4 (2014), 767–799

    2014

  73. [73]

    Deepesh V Hada and Shirish K Shevade. 2021. Rexplug: Explainable recommendation using plug-and-play language model. InProceedings of the 44th international ACM SIGIR conference on research and development in information retrieval. 81–91

    2021

  74. [74]

    F Maxwell Harper and Joseph A Konstan. 2015. The movielens datasets: History and context.Acm transactions on interactive intelligent systems (tiis)5, 4 (2015), 1–19

    2015

  75. [75]

    Matthew J Hausknecht and Peter Stone. 2015. Deep Recurrent Q-Learning for Partially Observable MDPs.. InAAAI fall symposia, Vol. 45. 141

    2015

  76. [76]

    Shirley Anugrah Hayati, Dongyeop Kang, Qingxiaoyang Zhu, Weiyan Shi, and Zhou Yu. 2020. Inspired: Toward sociable recommendation dialog systems. InProceedings of the 2020 conference on empirical methods in natural language processing (EMNLP). 8142–8152

    2020

  77. [77]

    Xiangnan He, Kuan Deng, Xiang Wang, Yan Li, Yongdong Zhang, and Meng Wang. 2020. Lightgcn: Simplifying and powering graph convolution network for recommendation. InProceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval. Manuscript submitted to ACM 38 Bohao Wang et al. 639–648

    2020

  78. [78]

    Diana C Hernandez-Bocanegra, Tim Donkers, and Jürgen Ziegler. 2020. Effects of argumentative explanation types on the perception of review-based recommendations. InAdjunct Publication of the 28th ACM Conference on User Modeling, Adaptation and Personalization. 219–225

    2020

  79. [79]

    Balázs Hidasi, Alexandros Karatzoglou, Linas Baltrunas, and Domonkos Tikk. 2015. Session-based recommendations with recurrent neural networks.arXiv preprint arXiv:1511.06939(2015)

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  80. [80]

    Minjie Hong, Yan Xia, Zehan Wang, Jieming Zhu, Ye Wang, Sihang Cai, Xiaoda Yang, Quanyu Dai, Zhenhua Dong, Zhimeng Zhang, et al. 2025. EAGER-LLM: Enhancing Large Language Models as Recommenders through Exogenous Behavior-Semantic Integration. InProceedings of the ACM on Web Conference 2025. 2754–2762

    2025

Showing first 80 references.