SimAug: Enhancing Recommendation with Pretrained Language Models for Dense and Balanced Data Augmentation

Huiyuan Chen; Tyler Derr; Xiaodong Yang; Xiran Fan; Yiwei Cai; Yu Wang; Yuying Zhao

REVIEW 4 major objections 5 minor 41 references

SimAug: Enhancing Recommendation with Pretrained Language Models for Dense and Balanced Data Augmentation

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read A pre-processing pass that adds text-similar unpopular items to inactive users' histories improves both recommendation accuracy and item fairness in collaborative filtering.

desk verdict A plausible augmentation method with a consistent empirical payoff, but a missing train/test split description could undermine the headline gains if synthetic edges leak test positives into training. read the letter →

arxiv 2505.01695 v1 pith:JIKMAOTZ submitted 2025-05-03 cs.IR

Yuying Zhao , Xiaodong Yang , Huiyuan Chen , Xiran Fan , Yu Wang , Yiwei Cai , Tyler Derr This is my paper · ORCID

classification cs.IR

keywords dataaugmentationpretrainedlanguagemodelsrecommendersystemsfairnesspopularitybiascollaborativefilteringLightGCNitemsimilarity

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

SimAug is a pre-processing data augmentation method that adds edges between inactive users and textually similar unpopular items, using similarity in a pretrained-language-model embedding space of item titles. The paper claims that this simple addition makes the interaction graph denser and more balanced, and that training the standard LightGCN recommender on the augmented data improves both utility (Recall@20, NDCG@20, and the average of five ranking metrics) and item fairness (unpopular-to-popular performance ratio) across nine Amazon datasets. Compared with random augmentation, which degrades performance, and augmentation based on collaborative-filtering embeddings, which worsens unfairness, SimAug consistently improves both objectives. A sympathetic reader cares because the method is model-agnostic, requires no prompt engineering or new loss functions, and suggests that world knowledge in pretrained language models can counteract popularity bias rather than amplify it.

What carries the argument

The central object is the textual item-embedding space produced once by the sentence transformer all-MiniLM-L6-v2 (with other embedding and LLM models also explored). The mechanism is a two-step pre-processing pipeline: first, item titles are encoded into cached textual embeddings; second, for every inactive user, each historical item contributes its top-k most similar unpopular items by cosine similarity in that space, and K edges are randomly sampled from the union of these candidates and added to the training graph. The design deliberately restricts augmentation to the region where data is most scarce, and randomly samples within the similarity-ranked candidate pool to limit injected noise. The same hyperparameters are used for all backbone training, so any performance change is attributed to the augmented interaction data itself.

What would settle it

On a held-out temporal split of one large dataset, compute whether the top-k textually similar unpopular items of each inactive user's historical items are actually more likely to be in that user's future interactions than randomly chosen unpopular items of the same popularity. If the future-interaction hit rate of the similarity-selected items equals or falls below the random baseline, the central mechanism is not capturing preference; in that case a replication comparing SimAug against the same number of random unpopular-item edges should show no utility advantage, which would falsify the claim that PLM similarity drives the gains.

Watch

Extended reading notes

Core claim

The paper's central claim is that textual similarity from a pretrained language model is a sufficiently reliable source of knowledge for interaction augmentation: for each inactive user and each item they already interacted with, collecting the top-k most similar items among the unpopular set in the PLM embedding space and randomly adding K such edges yields a denser, more balanced dataset. Trained on this dataset, LightGCN achieves consistent utility gains over the vanilla model, with average-score improvements between roughly 5% and 40% across the nine datasets, and substantially larger fairness improvements, since unpopular-item utility rises far more than popular-item utility. The paper further shows that the mechanism matters: random insertion of the same number of edges hurts, and augmentation from recommendation embeddings exaggerates popularity bias, while augmenting based on item similarity outperforms augmenting based on user similarity and also outperforms feature-level integration of text embeddings. In the paper's framing, this isolates the effect of data quality, showing that the gains come from the augmented dataset itself rather than from architectural or objective changes.

Load-bearing premise

The load-bearing assumption is that textual similarity between item titles, computed by a pretrained language model, is a reliable proxy for user preference: someone who interacted with an item is likely to be interested in textually similar unpopular items, even though this is never checked against actual co-interaction behavior.

Editorial extensions

If this is right

Any ID-based collaborative-filtering model, not just LightGCN, can in principle benefit from the augmented dataset, since SimAug changes only the input data and places no constraints on the downstream architecture.
Fairness gains are driven by large relative improvements in unpopular-item recommendation, not by sacrificing popular-item performance, so the method targets the imbalance rather than trading off one group against another.
The choice of PLM is not critical: smaller embedding models match larger ones, so the augmentation pipeline can stay lightweight and computationally cheap.
The number of added interactions is a tunable knob with an optimum; past that point, extra similar-item edges degrade top-ranked recommendation quality, so the method is not simply 'more data is better'.
Pre-processing augmentation of this sort is complementary to in-processing fairness methods, since it can be applied before any existing training loop.

Reading between the lines

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

The consistent gains may partly reflect that title similarity acts as a coarse category or brand signal rather than fine-grained preference; a variant using simple lexical overlap or category labels as the similarity source would test how much of the gain is specifically attributable to PLM world knowledge.
Because the intervention rebalances degrees by adding only unpopular-item edges, it could be viewed as a graph-structure rebalancing recipe that transfers beyond recommendation, for instance to node classification on text-attributed graphs where popularity is also skewed.
The paper does not validate the preference assumption against co-interaction patterns; a natural extension is to weight augmented edges by a confidence estimate rather than treating them as observed data.
If the assumption holds, the same recipe could be used with user-side augmentation on platforms that have richer user text, such as reviews or profiles, where the paper's user-based variant underperforms.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

SimAug is a simple, plausible pre-processing method: use a PLM's title embeddings to find textually similar unpopular items, and add those as training edges for inactive users. The empirical story is consistent across nine datasets, and the comparison against random augmentation and recommendation-embedding augmentation is a good control. The paper is honest about not yet understanding why it works. They also release code, which helps reproducibility.

The biggest problem is a missing methodological detail: the train/test split is never described. Algorithm 1 takes 'historical interactions H' as the input for augmentation, and the paper does not state whether H is the full interaction set or only the training split. If augmentation happens before splitting, then for an inactive user a held-out test item is exactly the kind of item SimAug may inject into training—it is unpopular, and its title similarity to a training item can place it in the candidate pool. That would directly inflate Recall@20 and NDCG@20 gains. This is not a minor omission; it determines whether the headline results mean anything. The authors need to state the split, apply augmentation only to the training portion, and ideally confirm no synthetic edge is also a test positive.

Other soft spots, in decreasing order: only LightGCN is tested despite the abstract claiming 'any systems'; three repeats without standard deviations are not enough to support the smaller gains (e.g., 5% Recall improvements on Pet Supplies); and the fairness gains are partly by construction, since only unpopular items receive extra edges and the Unpop/Pop ratio amplifies small absolute changes. The text-similarity-as-preference assumption is plausible but not directly validated—the Aug-Rec baseline helps, but it does not prove users actually like textually similar items.

That said, the core idea is reasonable and the paper could be solid after revision. It deserves a serious referee. I'd send it out, with a request that the authors clarify the split and add at least one more backbone and some significance tests.

Referee Report

4 major / 5 minor

Summary. The paper proposes SimAug, a pre-processing data-augmentation method for collaborative-filtering recommenders. SimAug first embeds item titles with a pretrained language model (all-MiniLM-L6-v2), then for each inactive user and each item in that user's history it identifies the top-k textually similar unpopular items and randomly selects K of them to add as synthetic interactions. The authors evaluate training LightGCN on nine Amazon datasets with the augmented data and report consistent improvements over the vanilla model, over a random augmentation baseline, and over an augmentation baseline based on pretrained recommendation embeddings, in both utility metrics (Recall@20, NDCG@20, Avg@20) and an item-fairness ratio (unpopular-item utility over popular-item utility). Additional experiments compare item- versus user-based augmentation, interaction- versus feature-based use of text, and several PLMs. The paper includes the code repository and a brief limitations section.

Significance. If the reported results hold, SimAug is an interesting and practical contribution: it is a lightweight, model-agnostic pre-processing step that converts textual item metadata into denser training data and appears to improve long-tail item performance without changing the downstream recommender. The evaluation covers nine public datasets and includes comparisons against other augmentation strategies, which is a genuine strength. The release of code also aids reproducibility. However, the significance is currently limited by several evaluation gaps, most importantly the complete absence of any description of the train/test split and of whether augmentation is applied before or after splitting; this makes the headline gains difficult to audit. The 'any systems' claim is also unsupported because only LightGCN is tested. If the authors can close these gaps, the paper would be a solid contribution to the data-augmentation-for-recommendation literature.

major comments (4)

[Section 3.2 / Algorithm 1 / Section 4.1.1] The manuscript never specifies how user-item interactions are divided into training, validation, and test sets, and it never states whether augmentation is applied to the training split only or to the full interaction set. Algorithm 1 takes 'historical interactions H' as the input to augmentation, and Section 3.2 adds edges between inactive users and unpopular items whose titles are textually similar to items in H. If H includes held-out test interactions, then for an inactive user a test positive (u,i) can be injected as a synthetic training edge whenever i is unpopular and textually similar to an item in u's history, inflating Recall@20 and NDCG@20 in Table 1 by target leakage. The presence of an early-stopping validation set in Appendix A.2 implies that some split exists, but the split rule and the position of augmentation relative to the split are not stated. This is a load-bearing methodological omission and should be fixed by explicitly describing the split and confirming (ideally with an experiment) that augmentation uses only training-split edges.
[Section 4.1.2 / Abstract] The abstract and introduction describe SimAug as a plug-and-play component that can be 'seamlessly integrated to any systems,' but Section 4.1.2 states that all experiments use LightGCN only, with no other backbone evaluated. The central plug-and-play claim is therefore not empirically supported. The authors should add at least one additional recommender backbone (for example a matrix-factorization model or a graph-based model other than LightGCN) to the main utility or fairness tables, or explicitly restrict the claim to LightGCN.
[Section 4.2 / Tables 1 and 3] The paper reports averages over three repetitions but omits standard deviations, saying only that they are 'often negligible.' Without standard deviations, confidence intervals, or significance tests, the reader cannot judge whether the improvements in Table 1 (e.g., +5.37% Recall for Pet Supplies, +3.67% NDCG for Toys and Games) are reliable or within run-to-run noise. The authors should report standard deviations or confidence intervals for the key comparisons, or provide a significance test such as a paired bootstrap or paired t-test across datasets.
[Section 3.2 / Section 4.3] The method's core assumption is that textual similarity of item titles, as computed by a PLM, is a reliable proxy for user preference: if a user interacted with an item, they are likely to be interested in textually similar unpopular items. This assumption is not validated against observed co-interaction patterns, nor is there any analysis of how often the augmented edges correspond to real user behavior. While the end-to-end improvement over Aug-Rec provides indirect evidence, the assumption is load-bearing for the method's rationale. The authors should include a direct validation, for example by measuring whether textually similar unpopular items are co-interacted more often than random unpopular items in held-out data, or by reporting the precision of augmented edges against a time-based held-out split.

minor comments (5)

[Algorithm 1, line 16] Line 16 uses RandomSample(Cu, k) but the input parameter is K; the pseudocode should use K to match the text of Section 3.2.
[Table 2 / Section 4.2] Section 4.2 says the fairness score is a ratio between 0 and 1, but Table 2 reports values such as 34.3659 and 3100.60%, which appear to be percentages. The unit should be stated explicitly and consistently.
[Section 4.6 / Table 3] Table 3 shows LLaMA2-13b on Office Products performing below the vanilla model (Recall@20 0.0469 vs 0.0485), which contradicts the statement that 'regardless of the specific PLM' performance improves consistently; this exception should be acknowledged and discussed.
[Section 4.4] The user-based augmentation variant is described in prose but no pseudocode or algorithm is provided, which makes it harder to reproduce; adding a short algorithm or pseudo-code would help.
[Section 4.3.2] The fairness evaluation excludes Aug-Random with the justification that its utility is poor, but no fairness numbers for Aug-Random are reported; including them, even in an appendix, would make the comparison complete.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: SimAug is an empirical augmentation method with no derivation that reduces to its inputs.

full rationale

The paper does not contain a formal derivation chain; its central claim is an empirical evaluation of a data augmentation procedure across nine datasets. The fairness improvement is a direct effect of augmenting interactions only for unpopular items, which is the method's explicitly stated mechanism rather than a hidden circular step. Utility improvements are contingent on the unvalidated assumption that PLM textual similarity tracks user preference; that is a correctness risk, not a circularity. No parameter is fitted to the evaluation metrics, no uniqueness theorem is imported from the authors' prior work, and self-citations are used only as background support for general claims (e.g., that biased data leads to biased models). The unspecified train/test split could create leakage, but that is a methodological concern outside the circularity definition. Therefore, as far as circularity is concerned, the paper's claims are self-contained and no result reduces to its inputs by construction.

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

No new entities are posited. The method relies on free parameters (k, K, user activity threshold) that are not reported, and the core assumption that PLM title similarity reflects user preference is load-bearing but unvalidated.

free parameters (3)

k (top-k similar items) = not reported
Number of nearest unpopular items considered per interacted item; used in Algorithm 1 step 13; value not specified in the paper.
K (augmented items per user) = not reported
Number of augmented interactions sampled per inactive user; line 16 of Algorithm 1 uses k instead of K, and the value is not reported. The paper shows performance varies with this in Fig. 5.
active/inactive user threshold = not reported
Partition of users into active and inactive follows Li et al. (2021) but the exact threshold is not given, affecting who gets augmented.

assumptions (3)

domain assumption Item title embeddings from PLMs capture semantic item similarity relevant to user taste.
Core of the augmentation mechanism; Section 3.2 builds all augmented edges on this similarity signal.
domain assumption Adding interactions to unpopular items improves their recommendations without harming overall utility.
Assumes the synthetic edges are positive signals; the paper shows performance degrades with too many augmentations, indicating the assumption holds only up to a point.
domain assumption Items without titles can be excluded without biasing the evaluation.
Pre-processing filters out titleless items; this could exclude items with poor metadata and affect the reported gains.

how reviews work

0 comments

Cite this review

Pith. "Pith review of SimAug: Enhancing Recommendation with Pretrained Language Models for Dense and Balanced Data Augmentation." pith.science (2026). https://pith.science/paper/JIKMAOTZ

@misc{pith2026250501695,
  author       = {Pith},
  title        = {Pith review of: SimAug: Enhancing Recommendation with Pretrained Language Models for Dense and Balanced Data Augmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JIKMAOTZ}},
  note         = {Machine review of arXiv:2505.01695}
}

read the original abstract

Deep Neural Networks (DNNs) are extensively used in collaborative filtering due to their impressive effectiveness. These systems depend on interaction data to learn user and item embeddings that are crucial for recommendations. However, the data often suffers from sparsity and imbalance issues: limited observations of user-item interactions can result in sub-optimal performance, and a predominance of interactions with popular items may introduce recommendation bias. To address these challenges, we employ Pretrained Language Models (PLMs) to enhance the interaction data with textual information, leading to a denser and more balanced dataset. Specifically, we propose a simple yet effective data augmentation method (SimAug) based on the textual similarity from PLMs, which can be seamlessly integrated to any systems as a lightweight, plug-and-play component in the pre-processing stage. Our experiments across nine datasets consistently demonstrate improvements in both utility and fairness when training with the augmented data generated by SimAug. The code is available at https://github.com/YuyingZhao/SimAug.

Figures

Figures reproduced from arXiv: 2505.01695 by the authors.

**Figure 1.** Two-step SimAug framework for interaction augmentation based on Pretrained Language Models (PLMs). The knowledge from PLM is first extracted in the textual embeddings and then interactions are augmented based on the embeddings. 2 Data Sparsity and Imbalance In this section, we empirically investigate the relationship between sparsity/density and utility performance, and the relationship between data imbalance an… view at source ↗

**Figure 2.** Investigating data sparsity and imbalance is [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 5.** Impact of the number of augmentation. and General Text Embeddings (GTE) model (Li et al., 2023)), we directly obtained the textual embeddings. For large language models (LLMs) that are not tailored for embedding task, we followed (Jiang et al., 2023) to extract their embeddings. The results, shown in [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Comparison of other content-based variants. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 23 canonical work pages

[1]

online" 'onlinestring :=

ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRING...
[2]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...
[3]

Himan Abdollahpouri, Robin Burke, and Bamshad Mobasher. 2019. Managing popularity bias in recommender systems with personalized re-ranking. arXiv preprint arXiv:1901.07555

arXiv 2019
[4]

Abdul Basit Ahanger, Syed Wajid Aalam, Muzafar Rasool Bhat, and Assif Assad. 2022. Popularity bias in recommender systems-a review. In International Conference on Emerging Technologies in Computer Engineering, pages 431--444. Springer

work page 2022
[5]

Nadia F Al-Bakri and Soukaena Hassan Hashim. 2018. Reducing data sparsity in recommender systems. Al-Nahrain Journal of Science, 21(2):138--147

work page 2018
[6]

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. In Proceedings of the 17th ACM Conference on Recommender Systems, pages 1007--1014

2023
[7]

Tom B Brown. 2020. Language models are few-shot learners. arXiv preprint arXiv:2005.14165

arXiv 2020
[8]

Huiyuan Chen and Jing Li. 2019. Adversarial tensor factorization for context-aware recommendation. In Proceedings of the 13th ACM Conference on Recommender Systems, pages 363--367

work page 2019

Show all 41 references

[9]

Huiyuan Chen, Yusan Lin, Menghai Pan, Lan Wang, Chin-Chia Michael Yeh, Xiaoting Li, Yan Zheng, Fei Wang, and Hao Yang. 2022. Denoising self-attentive sequential recommendation. In Proceedings of the 16th ACM conference on recommender systems

2022
[10]

Rui Chen, Qingyi Hua, Yan-Shuo Chang, Bo Wang, Lei Zhang, and Xiangjie Kong. 2018. A survey of collaborative filtering-based recommender systems: From traditional methods to hybrid methods based on social networks. IEEE access, 6:64301--64320

2018
[11]

Zhihong Chen, Rong Xiao, Chenliang Li, Gangfeng Ye, Haochuan Sun, and Hongbo Deng. 2020. Esam: Discriminative domain adaptation with non-displayed items to improve long-tail performance. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development ...

2020
[12]

Gallegos, Ryan A

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 Linguistics

2024
[13]

Muhammad Usman Hadi, Rizwan Qureshi, Abbas Shah, Muhammad Irfan, Anas Zafar, Muhammad Bilal Shaikh, Naveed Akhtar, Jia Wu, Seyedali Mirjalili, et al. 2023. A survey on large language models: Applications, challenges, limitations, and practical usage. Authorea Preprints

2023
[14]

Xiangnan He, Kuan Deng, Xiang Wang, Yan Li, Yongdong Zhang, and Meng Wang. 2020. Lightgcn: Simplifying and powering graph convolution network for recommendation. In Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval,...

2020
[15]

Yupeng Hou, Jiacheng Li, Zhankui He, An Yan, Xiusi Chen, and Julian McAuley. 2024. Bridging language and items for retrieval and recommendation. arXiv preprint arXiv:2403.03952

2024 arXiv
[16]

Feiran Huang, Zhenghang Yang, Junyi Jiang, Yuanchen Bei, Yijie Zhang, and Hao Chen. 2024. Large language model interaction simulator for cold-start item recommendation. arXiv preprint arXiv:2402.09176

2024 arXiv
[17]

Jiayuan Huang, Arthur Gretton, Karsten Borgwardt, Bernhard Sch \"o lkopf, and Alex Smola. 2006. Correcting sample selection bias by unlabeled data. Advances in neural information processing systems, 19

2006
[18]

Ting Jiang, Shaohan Huang, Zhongzhi Luan, Deqing Wang, and Fuzhen Zhuang. 2023. Scaling sentence embeddings with large language models. arXiv preprint arXiv:2307.16645

2023 arXiv
[19]

Anastasiia Klimashevskaia, Dietmar Jannach, Mehdi Elahi, and Christoph Trattner. 2024. A survey on popularity bias in recommender systems. User Modeling and User-Adapted Interaction, pages 1--58

2024
[20]

Hyeyoung Ko, Suyeon Lee, Yoonseo Park, and Anna Choi. 2022. A survey of recommendation systems: recommendation models, techniques, and application fields. Electronics, 11(1):141

2022
[21]

Vivian Lai, Huiyuan Chen, Chin-Chia Michael Yeh, Minghua Xu, Yiwei Cai, and Hao Yang. 2023. Enhancing transformers without self-supervised learning: A loss landscape perspective in sequential recommendation. In Proceedings of the 17th ACM Conference on Recommender Systems

2023
[22]

Yunqi Li, Hanxiong Chen, Zuohui Fu, Yingqiang Ge, and Yongfeng Zhang. 2021. User-oriented fairness in recommendation. In Proceedings of the web conference 2021, pages 624--632

2021
[23]

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

2023 arXiv
[24]

Jianghao Lin, Xinyi Dai, Yunjia Xi, Weiwen Liu, Bo Chen, Hao Zhang, Yong Liu, Chuhan Wu, Xiangyang Li, Chenxu Zhu, et al. 2023. How can recommender systems benefit from large language models: A survey. arXiv preprint arXiv:2306.05817

2023 arXiv
[25]

Kelong Mao, Jieming Zhu, Xi Xiao, Biao Lu, Zhaowei Wang, and Xiuqiang He. 2021. Ultragcn: ultra simplification of graph convolutional networks for recommendation. In Proceedings of the 30th ACM international conference on information & knowledge management, pages 1253--1262

2021
[26]

Zhaopeng Qiu, Xian Wu, Jingyue Gao, and Wei Fan. 2021. U-bert: Pre-training user representations for improved recommendation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 4320--4327

2021
[27]

Leheng Sheng, An Zhang, Yi Zhang, Yuxin Chen, Xiang Wang, and Tat-Seng Chua. 2025. Language representations can be what recommenders need: Findings and potentials. In The Thirteenth International Conference on Learning Representations

2025
[28]

Zuoli Tang, Zhaoxin Huan, Zihao Li, Xiaolu Zhang, Jun Hu, Chilin Fu, Jun Zhou, and Chenliang Li. 2023. One model for all: Large language models are domain-agnostic recommendation systems. arXiv preprint arXiv:2310.14304

2023 arXiv
[29]

Fadi Thabtah, Suhel Hammoud, Firuz Kamalov, and Amanda Gonsalves. 2020. Data imbalance in classification: Experimental evaluation. Information Sciences, 513:429--441

2020
[30]

Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timoth \'e e Lacroix, Baptiste Rozi \`e re, Naman Goyal, Eric Hambro, Faisal Azhar, et al. 2023. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971

2023 arXiv
[31]

Jianling Wang, Haokai Lu, James Caverlee, Ed H Chi, and Minmin Chen. 2024 a . Large language models as data augmenters for cold-start item recommendation. In Companion Proceedings of the ACM on Web Conference 2024, pages 726--729

2024
[32]

Rossi, Namyong Park, Huiyuan Chen, Nesreen K

Yu Wang, Ryan A. Rossi, Namyong Park, Huiyuan Chen, Nesreen K. Ahmed, Puja Trivedi, Franck Dernoncourt, Danai Koutra, and Tyler Derr. 2025. A large-scale training paradigm for graph generative models. In The Thirteenth International Conference on Learning Representations

2025
[33]

Yu Wang, Yuying Zhao, Yushun Dong, Huiyuan Chen, Jundong Li, and Tyler Derr. 2022. Improving fairness in graph neural networks via mitigating sensitive attribute leakage. In Proceedings of the 28th ACM SIGKDD conference on knowledge discovery and data mining

2022
[34]

Yuan Wang, Xuyang Wu, Hsin-Tai Wu, Zhiqiang Tao, and Yi Fang. 2024 b . Do large language models rank fairly? an empirical study on the fairness of LLM s as rankers. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguis...

2024
[35]

Junda Wu, Cheng-Chun Chang, Tong Yu, Zhankui He, Jianing Wang, Yupeng Hou, and Julian McAuley. 2024 a . Coral: Collaborative retrieval-augmented large language models improve long-tail recommendation. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and ...

2024
[36]

Likang Wu, Zhi Zheng, Zhaopeng Qiu, Hao Wang, Hongchao Gu, Tingjia Shen, Chuan Qin, Chen Zhu, Hengshu Zhu, Qi Liu, et al. 2024 b . A survey on large language models for recommendation. World Wide Web, 27(5):60

2024
[37]

Xiaodong Yang, Huiyuan Chen, Yuchen Yan, Yuxin Tang, Yuying Zhao, Eric Xu, Yiwei Cai, and Hanghang Tong. 2024. Simce: Simplifying cross-entropy loss for collaborative filtering. arXiv preprint arXiv:2406.16170

2024 arXiv
[38]

Zheng Yuan, Fajie Yuan, Yu Song, Youhua Li, Junchen Fu, Fei Yang, Yunzhu Pan, and Yongxin Ni. 2023. Where to go next for recommender systems? id-vs. modality-based recommender models revisited. In Proceedings of the 46th International ACM SIGIR Conference on Research and Devel...

2023
[39]

Zhipeng Zhang, Yuhang Zhang, Mianxiong Dong, Kaoru Ota, Yao Zhang, and Yonggong Ren. 2024. Collaborative tag-aware graph neural network for long-tail service recommendation. IEEE Transactions on Services Computing

2024
[40]

Yuying Zhao, Minghua Xu, Huiyuan Chen, Yuzhong Chen, Yiwei Cai, Rashidul Islam, Yu Wang, and Tyler Derr. 2024. Can one embedding fit all? a multi-interest learning paradigm towards improving user interest diversity fairness. In Proceedings of the ACM Web Conference 2024

2024
[41]

Ziwei Zhu, Yun He, Xing Zhao, and James Caverlee. 2021. Popularity bias in dynamic recommendation. In Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining, pages 2439--2449

2021

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.

[1] [1]

online" 'onlinestring :=

ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRING...

[2] [2]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

[3] [3]

Himan Abdollahpouri, Robin Burke, and Bamshad Mobasher. 2019. Managing popularity bias in recommender systems with personalized re-ranking. arXiv preprint arXiv:1901.07555

arXiv 2019

[4] [4]

Abdul Basit Ahanger, Syed Wajid Aalam, Muzafar Rasool Bhat, and Assif Assad. 2022. Popularity bias in recommender systems-a review. In International Conference on Emerging Technologies in Computer Engineering, pages 431--444. Springer

work page 2022

[5] [5]

Nadia F Al-Bakri and Soukaena Hassan Hashim. 2018. Reducing data sparsity in recommender systems. Al-Nahrain Journal of Science, 21(2):138--147

work page 2018

[6] [6]

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. In Proceedings of the 17th ACM Conference on Recommender Systems, pages 1007--1014

2023

[7] [7]

Tom B Brown. 2020. Language models are few-shot learners. arXiv preprint arXiv:2005.14165

arXiv 2020

[8] [8]

Huiyuan Chen and Jing Li. 2019. Adversarial tensor factorization for context-aware recommendation. In Proceedings of the 13th ACM Conference on Recommender Systems, pages 363--367

work page 2019