{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T11:24:29Z","timestamp":1786965869645,"version":"build-2736575974"},"reference-count":28,"publisher":"Association for Computing Machinery (ACM)","issue":"3","funder":[{"name":"Coordena o de Aperfei oamento de Pessoal de N vel Superior","award":["Grant 001"],"award-info":[{"award-number":["Grant 001"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Interact. Intell. Syst."],"published-print":{"date-parts":[[2026,9,30]]},"abstract":"<jats:p>\n                    The adoption of Recommender Systems (RSs) in various domains has become increasingly popular, but concerns have been raised about their lack of transparency and interpretability. While significant advancements have been made in creating explainable RSs, there is still a shortage of automated approaches that can deliver meaningful and contextual human-centered explanations. Numerous studies have evaluated explanations based on human-generated recommendations and explanations to address this gap. However, such approaches do not scale for real-world systems. Building on recent research that exploits Large Language Models (LLMs) for RSs, we propose leveraging the conversational capabilities of ChatGPT to provide users with personalized, human-like, and meaningful explanations for recommended items. Our article presents a user study with 94 participants that measures users\u2019 perceptions of ChatGPT-generated explanations when acting as a recommender system. Regarding recommendations, we assess whether users prefer ChatGPT over random (but popular) recommendations. Concerning explanations, we assess users\u2019 perceptions of personalization, effectiveness, and persuasiveness. We also break down the explanations in its constituent arguments and investigate the differences in argument types between generic and user-specific explanations. Our results show that participants rated ChatGPT\u2019s recommendations significantly higher than random ones (\n                    <jats:inline-formula content-type=\"math\/tex\">\n                      <jats:tex-math notation=\"LaTeX\" version=\"MathJax\">\\(\\beta=-.53\\)<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    ,\n                    <jats:inline-formula content-type=\"math\/tex\">\n                      <jats:tex-math notation=\"LaTeX\" version=\"MathJax\">\\({\\textrm{p}} &lt; .001\\)<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    ). Surprisingly, user-based explanations that explicitly referenced participants\u2019 preferences were not perceived as more personalized or persuasive than generic ones, except when the recommended movie was unfamiliar, where user-based explanations became more effective (\n                    <jats:inline-formula content-type=\"math\/tex\">\n                      <jats:tex-math notation=\"LaTeX\" version=\"MathJax\">\\(\\beta=0.35\\)<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    ,\n                    <jats:inline-formula content-type=\"math\/tex\">\n                      <jats:tex-math notation=\"LaTeX\" version=\"MathJax\">\\({\\textrm{p}} &lt; .05\\)<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    ). Overall, our findings highlight both the promise and the limitations of using LLMs for generating personalized explanations, suggesting that personalization is most impactful when users lack prior knowledge of the item.\n                  <\/jats:p>","DOI":"10.1145\/3779059","type":"journal-article","created":{"date-parts":[[2025,12,3]],"date-time":"2025-12-03T12:37:18Z","timestamp":1764765438000},"page":"1-27","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["User Perceptions of Personalized and Generic Explanations in LLM-Driven Recommender Systems"],"prefix":"10.1145","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-1507-0580","authenticated-orcid":false,"given":"\u00cdtallo","family":"Silva","sequence":"first","affiliation":[{"name":"DSC, Federal University of Campina Grande, Campina Grande, Brazil"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7599-372X","authenticated-orcid":false,"given":"Leandro","family":"Balby","sequence":"additional","affiliation":[{"name":"DSC, Federal University of Campina Grande, Campina Grande, Brazil"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2929-0529","authenticated-orcid":false,"given":"Alan","family":"Said","sequence":"additional","affiliation":[{"name":"University of Gothenburg, Gothenburg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5908-9511","authenticated-orcid":false,"given":"Martijn C.","family":"Willemsen","sequence":"additional","affiliation":[{"name":"School of Innovation Sciences, Eindhoven University of Technology, Eindhoven, Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,8,17]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1145\/3397271.3401032"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1145\/2959100.2959153"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.565"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1145\/3690655"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0279720"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.23919\/MIPRO.2018.8400040"},{"key":"e_1_3_2_8_2","unstructured":"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:2303.14524. Retrieved from https:\/\/arxiv.org\/abs\/2303.14524"},{"key":"e_1_3_2_9_2","doi-asserted-by":"crossref","unstructured":"Rachel Harrison Anton Dereventsov and Anton Bibin. 2023. Zero-shot recommendations with pre-trained large language models for multimodal nudging. arXiv:2309.01026. Retrieved from https:\/\/arxiv.org\/abs\/2309.01026","DOI":"10.1109\/ICDMW60847.2023.00195"},{"key":"e_1_3_2_10_2","unstructured":"Xu Huang Jianxun Lian Yuxuan Lei Jing Yao Defu Lian and Xing Xie. 2023. Recommender AI agent: Integrating large language models for interactive recommendations. arXiv:2308.16505. Retrieved from https:\/\/arxiv.org\/abs\/2308.16505"},{"key":"e_1_3_2_11_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11257-011-9118-4"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1145\/3569423"},{"key":"e_1_3_2_13_2","unstructured":"Junling Liu Chao Liu Renjie Lv Kang Zhou and Yan Zhang. 2023. Is ChatGPT a good recommender? A preliminary study. arXiv:2304.10149. Retrieved from https:\/\/arxiv.org\/abs\/2304.10149"},{"key":"e_1_3_2_14_2","doi-asserted-by":"publisher","DOI":"10.1145\/3565480"},{"key":"e_1_3_2_15_2","doi-asserted-by":"publisher","DOI":"10.1145\/3631700.3665185"},{"key":"e_1_3_2_16_2","unstructured":"Hanjia Lyu Song Jiang Hanqing Zeng Qifan Wang Si Zhang Ren Chen Chris Leung Jiajie Tang Yinglong Xia and Jiebo Luo. 2023. LLM-Rec: Personalized recommendation via prompting large language models. arXiv:2307.15780. Retrieved from https:\/\/arxiv.org\/abs\/2307.15780"},{"key":"e_1_3_2_17_2","doi-asserted-by":"publisher","DOI":"10.1145\/1639714.1639731"},{"key":"e_1_3_2_18_2","unstructured":"Yubo Shu Hansu Gu Peng Zhang Haonan Zhang Tun Lu Dongsheng Li and Ning Gu. 2023. RAH! RecSys-assistant-human: A human-central recommendation framework with large language models. arXiv:2308.09904. Retrieved from https:\/\/arxiv.org\/abs\/2308.09904"},{"key":"e_1_3_2_19_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICDEW.2007.4401070"},{"key":"e_1_3_2_20_2","volume-title":"Argumentation, Objectivity and Bias: Proceedings of the Ontario Society for the Study of Argumentation Conference","volume":"11","author":"Wagemans J. H. M.","year":"2016","unstructured":"J. H. M. Wagemans. 2016. Constructing a periodic table of arguments. In Argumentation, Objectivity and Bias: Proceedings of the Ontario Society for the Study of Argumentation Conference, Vol. 11. L. Benacquista and P. Bondy (Eds.), OSSA."},{"key":"e_1_3_2_21_2","unstructured":"J. H. M. Wagemans. 2025. Argument Type Identification Procedure (ATIP)\u2014Version 5. Retrieved from www.periodic-table-of-arguments.org\/argument-type-identification-procedure"},{"key":"e_1_3_2_22_2","unstructured":"J. H. M. Wagemans. 2021. Periodic Table of Arguments 2.5 A. Retrieved from https:\/\/periodic-table-of-arguments.org\/about\/"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.resuscitation.2024.110404"},{"key":"e_1_3_2_24_2","unstructured":"Yan Wang Zhixuan Chu Xin Ouyang Simeng Wang Hongyan Hao Yue Shen Jinjie Gu Siqiao Xue James Y. Zhang Qing Cui et al. 2023. Enhancing recommender systems with large language model reasoning graphs. arXiv:2308.10835. Retrieved from https:\/\/arxiv.org\/abs\/2308.10835"},{"key":"e_1_3_2_25_2","unstructured":"Likang Wu Zhi Zheng Zhaopeng Qiu Hao Wang Hongchao Gu Tingjia Shen Chuan Qin Chen Zhu Hengshu Zhu Qi Liu et al. 2023. A survey on large language models for recommendation. arXiv:2305.19860. Retrieved from https:\/\/arxiv.org\/abs\/2305.19860"},{"key":"e_1_3_2_26_2","doi-asserted-by":"publisher","unstructured":"Yongfeng Zhang and Xu Chen. 2020. Explainable recommendation: A survey and new perspectives. Foundations and Trends\u00ae in Information Retrieval 14 1 (Mar. 2020) 1\u2013101. DOI: 10.1561\/1500000066","DOI":"10.1561\/1500000066"},{"key":"e_1_3_2_27_2","volume-title":"Proceedings on I (Still) Can\u2019t Believe It\u2019s Not Better! NeurIPS 2021 Workshop","author":"Zhang Yuhui","year":"2021","unstructured":"Yuhui Zhang, H. A. O. Ding, Zeren Shui, Yifei Ma, James Zou, Anoop Deoras, and Hao Wang. 2021. Language models as recommender systems: Evaluations and limitations. In Proceedings on I (Still) Can\u2019t Believe It\u2019s Not Better! NeurIPS 2021 Workshop. Retrieved from https:\/\/openreview.net\/forum?id=hFx3fY7-m9b"},{"key":"e_1_3_2_28_2","doi-asserted-by":"publisher","DOI":"10.1145\/2600428.2609579"},{"key":"e_1_3_2_29_2","volume-title":"Proceedings of the IntRS \u201923 Workshop","author":"Zhou Joyce","year":"2023","unstructured":"Joyce Zhou and Thorsten Joachims. 2023. GPT as a baseline for recommendation explanation texts. In Proceedings of the IntRS \u201923 Workshop."}],"container-title":["ACM Transactions on Interactive Intelligent Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3779059","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T10:28:38Z","timestamp":1786962518000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3779059"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,17]]},"references-count":28,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,9,30]]}},"alternative-id":["10.1145\/3779059"],"URL":"https:\/\/doi.org\/10.1145\/3779059","relation":{},"ISSN":["2160-6455","2160-6463"],"issn-type":[{"value":"2160-6455","type":"print"},{"value":"2160-6463","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,17]]},"assertion":[{"value":"2025-01-23","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-11-18","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-17","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}