Pith. sign in

REVIEW 4 major objections 6 minor 78 references

CORONA: A Coarse-to-Fine Framework for Graph-based Recommendation with Large Language Models

T0 review · 4 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 CORONA claims that using LLM preference and intent reasoning to prune the interaction graph before GNN scoring lifts recommendation accuracy, with average relative gains around 18% in recall and NDCG.

desk verdict CORONA's real novelty is putting LLM reasoning inside candidate filtering, but a circular training dependency and an abstract/body numbers mismatch need fixing. read the letter →

arxiv 2506.17281 v1 pith:RWDXVAKJ submitted 2025-06-14 cs.IR cs.AI

Junze Chen , Xinjie Yang , Cheng Yang , Junfei Bao , Zeyuan Guo , Yawen Li , Chuan Shi This is my paper
classification cs.IRcs.AI
keywords largelanguagemodelsgraphneuralnetworksrecommendersystemscandidatefilteringcoarse-to-fineretrievalcollaborativesubgraphcold-startrecommendation
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

CORONA is a recommendation framework that tries to use large language models not for reranking or data augmentation, but inside the candidate-filtering step itself. The paper's central claim is that two rounds of LLM reasoning—first about the user's general preferences, then about their short-term intent—can be encoded as query embeddings that progressively prune the user–item interaction graph, and that a graph neural network scoring the surviving subgraph recommends better than previous methods. On Netflix, MovieLens, and Amazon-book, the authors report average relative gains over the best baseline; the abstract gives 18.6% for recall and 18.4% for NDCG, while the main-results table reports 17.66% and 16.06%. If the claim holds, the practical consequence is that LLM world knowledge and graph collaborative filtering complement each other inside retrieval rather than competing only at the final ranking step.

What carries the argument

The central object is the subgraph retriever, which compares an LLM-generated query embedding against user embeddings built from raw user features plus a distance encoding. Distance encoding is a small learned vector that tells the retriever whether another user is one hop, two hops, or more than two hops away from the target user in the interaction graph, and a shared linear layer projects the concatenated features into the same space as the query embedding. Cosine similarity then selects the top-$k$ users, and the items adjacent to those users are added to the subgraph. This happens twice, once for preference and once for intent, before a trainable GNN scores the final subgraph with inner products and a Bayesian Personalized Ranking loss. The retriever is the mechanism that injects LLM knowledge into the filtering stage, and the distance encoding is what keeps the filter collaborative rather than purely semantic.

What would settle it

Replace the LLM-generated query embeddings in the two retrieval steps with random vectors of the same dimension, keeping the subgraph sizes fixed, and retrain the GNN: if recall at rank 20 and NDCG at rank 20 on Netflix, MovieLens, and Amazon-book do not drop substantially, then the reported gains are not caused by LLM preference and intent reasoning.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that an LLM's free-text reasoning, encoded as a query embedding, can act as a coarse-grained filter over an interaction graph, and that the subgraph it selects is a better input to a GNN than the full graph, a fixed 1-hop neighborhood, or a fixed 2-hop neighborhood. The preference-assisted retrieval step uses the LLM's reading of a user profile to find similar users and their items; the intent-assisted step uses the user's purchase history plus a statistical summary of the first subgraph to narrow further; the final subgraph is scored by a GNN. The authors interpret their ablation study as showing that both reasoning stages are necessary, that the intent stage contributes more than the preference stage, and that the LLM's reasoning content—not just the user text—is doing the work. The paper also claims the framework remains effective for cold-start items and that the retrieved subgraph plus GNN gives the best results on all three datasets.

Load-bearing premise

The whole scheme depends on the idea that cosine similarity between the frozen LLM's text-embedding space and the learned projection of user features is a faithful measure of similar taste, but the paper never directly verifies that the two spaces are aligned.

Editorial extensions

If this is right

  • LLM reasoning can be a retrieval-stage operator over an interaction graph, not just a post-filter reranker or a dataset augmenter.
  • Both retrieval stages are load-bearing: the ablation shows removing either hurts, and removing the intent stage hurts more than removing the preference stage.
  • The LLM-guided subgraph, rather than the full graph or a fixed-hop neighborhood, is the right input for the GNN: under the same GNN, LLM-based subgraph retrieval beats full-graph, fixed 1-hop, and fixed 2-hop variants.
  • The advantage persists for cold-start items, which suggests the LLM's textual reasoning can substitute for missing interaction data when few users have rated an item.
  • The framework is compatible with different GNN backbones, so existing graph recommenders could adopt the two retrieval stages without replacing their scoring model.

Reading between the lines

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

  • A likely hidden component of the gain is the LLM's semantic priors about genres, directors, and item attributes; because the query encoder is frozen and only user-side features are trained, swapping in a model without world knowledge would isolate how much of the improvement is due to graph structure versus language priors.
  • The summary-and-statistics step compresses the retrieved subgraph into the top 20 attribute frequencies, which is a narrow bottleneck; richer structured summaries or multi-hop intent chaining could extend the same coarse-to-fine idea to catalogs with long-tail categories.
  • The distance-aware retriever assumes the interaction graph has enough two-hop overlap among similar users; on sparser graphs, the optimal subgraph size would likely need to grow or the distance encoding would need a learned threshold.
  • The reported subgraph-size optimum and the per-user cost estimate suggest an industrial deployment would tune the subgraph size against a latency budget rather than use the same value across catalogs.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes CORONA, a graph-based recommendation framework that interleaves LLM reasoning with graph retrieval. In Stage 1, an LLM reasons about a user's general preferences from a profile; the resulting text is embedded and used to retrieve the top-k users most similar to the target user via cosine similarity over learned projections of user features plus distance encodings. The items connected to these users are summarized. In Stage 2, the LLM is prompted with the user's history and this summary to infer a short-term intent, which again retrieves a smaller subgraph. In Stage 3, a GNN (GCN or GraphTransformer) embeds the target user on the subgraph and scores candidate items by inner product. The authors evaluate on Netflix, MovieLens-10M, and Amazon-book under top-N recall/NDCG, including cold-start, ablations, hyperparameter, efficiency, and case studies, and report large relative improvements over baselines.

Significance. Should the reported gains survive a fair comparison, CORONA provides a useful answer to a real question: how to inject LLM reasoning into the candidate-generation stage of graph recommendation rather than only re-ranking or augmenting. The framework is modular, and the ablation set in Tables 3 and 4 is informative, especially the comparison with fixed-hop subgraphs and with a second GNN backbone. The authors also report per-user inference cost, a practical consideration often omitted. The code availability is a plus. However, the manuscript's central claims currently rest on several points that need clarification or correction: the headline improvement numbers are internally inconsistent, the training objective in Eq. (4) has an unstated dependency, the evaluation protocol may be asymmetric between CORONA and the baselines, and no variance information is reported for the five-run averages.

major comments (4)
  1. [Abstract; §4.2, Table 1] The Abstract's headline numbers (18.6% recall, 18.4% NDCG) do not match Section 4.2's stated averages (17.66% recall, 16.06% NDCG), and Section 4.2's numbers are themselves exactly the averages of the nine Recall and nine NDCG relative-improvement values in Table 1's 'Improv.' row. Please reconcile the Abstract with the table and state precisely how the averages are computed (per metric across datasets, or per dataset across metrics). As it stands, the central claim in the Abstract is not supported by the reported table.
  2. [§3.4, Eq. (4)] The training loss L1 is not well-defined as stated because E_Q2 depends on θ. Equation (3) uses E_Q2 = Encode(LLM_IR(Summary(V'_1), L_u)), and V'_1 is obtained via Eqs. (1)-(2) using X_u' = Linear_θ(...); hence E_Q2 changes when θ changes. The paper does not say whether E_Q2 is precomputed once with an initial or untrained retriever, recomputed at each training step, or detached. This matters for the attribution: if E_Q2 is precomputed, then the 'intent-assisted' stage is not conditioned on the learned preference retrieval; if it is recomputed, the objective in Eq. (4) is a moving target and the ablations in Table 4 ('w/o Intent-assisted Retrieval', 'w/o Preference-assisted Retrieval') may compare against different protocols. Please specify the exact training protocol and, if necessary, re-run ablations under a fixed protocol.
  3. [§3.5, Eq. (5); §4.1.2; Table 1] There is an apparent mismatch between the evaluation protocol and the model's candidate set. Section 3.5 scores only items in the retrieved subgraph G', while Section 4.1.2 states that an 'all-ranking' strategy is employed, which in the cited baselines (e.g., LightGCN, RLMRec) normally means ranking all items. If CORONA is evaluated only over G' while baselines rank over the full item pool, the reported relative improvements in Table 1 could be inflated by the reduced candidate space rather than by better ranking quality. Please state explicitly which candidate set is used for each method and provide a comparison in which all methods are evaluated over the same candidate set (or CORONA is evaluated over all items); the 'GCN (full graph)' vs 'GCN (CORONA)' rows of Table 3 are relevant but do not resolve the question because they compare CORONA's subgraph against the full graph under the same GNN, not against the main baselines.
  4. [§4.1.2, Tables 1-4] The authors report averages of five independent runs but provide no standard deviations, confidence intervals, or significance tests. Several of the headline improvements are driven by single metrics (e.g., Netflix N@10 improvement is 2.95% in Table 1, whereas MovieLens N@10 is 30.96%), so without variance information it is impossible to judge whether the SOTA claim is robust. Please report per-run standard deviations for at least Tables 1 and 2, and state whether the differences over the best baseline are statistically significant (e.g., paired bootstrap or t-test across runs).
minor comments (6)
  1. [§1] The sentence 'The contribution of this work are three-fold' should be 'The contributions of this work are three-fold' or 'The contribution of this work is three-fold'.
  2. [Table 2] In the MovieLens column, BinLLM's R@10 value appears as 0.01373, which is likely a typo for 0.1373 given the other values; please check and correct.
  3. [§3.3.2] The phrase 'very few users in Ztrain pay attention to items interacted by users beyond two-hop neighbors' is unclear; the statistic is also reported for Netflix and MovieLens only, so it should be stated more carefully rather than as a general property.
  4. [§4.1.4] The implementation details say user/item features are obtained with BERT while query embeddings use 'text-embedding-ada-002'; please clarify whether the 128-dimensional user and item features are also from the same embedding family or from a separate BERT encoder.
  5. [§4.7, Figure 6] The two case panels in Figure 6 appear to be duplicated in the submitted text; please check the figure layout so that each panel matches its (a)/(b) label and the corresponding user profile.
  6. [§3.4] The intent-reasoning prompt template includes 'Candidate summary' in the instruction, but the example does not show the actual candidate-summary placeholders; align the prompt template with the Summary & Statistics output from Section 3.3.3.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the LLM-guided retrieval and GNN ranking are trained on external data with distinct losses; the only coupling (E_Q2 depends on θ through the Stage-1 subgraph) is a training-protocol ambiguity, not a definitional reduction.

full rationale

The derivation chain is self-contained. The LLM queries in Eqs. 2-3 are produced by a frozen external model from user profiles, histories, and candidate summaries; the retriever parameters θ and distance encodings are trained via Eq. 4 against ground-truth interaction users; and the final ranking comes from a separately trained GNN with BPR loss (Eqs. 5-6) on the progressively pruned subgraph. I checked for self-citation load-bearing claims (all cited baselines are external and none of the authors' own prior work is invoked to force a design choice), imported uniqueness theorems (none), ansatz-by-citation (the cosine-similarity retrieval and distance encoding are introduced in this paper, not inherited), and renaming of a known result (the coarse-to-fine LLM retrieval plus GNN pipeline is not a relabeling of an existing method). The one notable dependency is that Eq. 4 uses E_Q2, which is generated from a subgraph produced by Eq. 2 using the same θ being optimized; the paper does not state whether E_Q2 is recomputed during training. This is a training-protocol ambiguity and a potential implementation risk, but it is not a definitional circularity: E_Q2 is an LLM-encoded text summary rather than a model parameter, and the final GNN scores are not equivalent to the retriever's inputs by construction. The central claims are benchmarked against external datasets and baselines, so no circularity beyond 0 is warranted.

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

The method relies on several domain assumptions about the alignment between LLM embedding spaces and interaction-graph features, and on the sufficiency of two-hop neighborhoods. The free parameters listed (subgraph sizes, temperature, distance encoding dimension, GNN hidden dimension) were tuned on the same datasets used for evaluation.

free parameters (6)
  • Retrieved subgraph size k = 3000 (best)
    The number of most similar users selected in preference-assisted retrieval; tuned per dataset via R@20 (Section 4.5.1). Performance peaks around k=3000.
  • Intent retrieval size k2 = Not reported
    The number of users selected in the second-stage retrieval (Eq. 3); no explicit value given, likely smaller than k.
  • Distance encoding dimension dim(e) = 2
    The dimension of the per-hop distance embeddings; tuned over {0..5} with best at 2 (Section 4.5.3).
  • LLM temperature tau = Not explicitly stated (range 0-1 tested)
    Controls randomness of LLM generation; best performance at low values, but exact best not reported (Section 4.5.2).
  • GNN hidden dimension = 128
    Hidden dimension for the two-layer GCN; tuned over {8..256} with best at 128 (Section 4.5.4).
  • Negative sample size = 10
    Number of negative items sampled for the BPR loss (Eq. 6); set in implementation details.
assumptions (4)
  • domain assumption Users within two hops of the target user capture most of the relevant collaborative signal
    Justifies assigning a single distance encoding e3 to all users beyond two hops (Section 3.3.2, based on an analysis that only 3% of Netflix and 0.4% of MovieLens users interact with items beyond two hops).
  • domain assumption An LLM-generated preference/intent description, when encoded by a fixed embedding model, produces a query vector that is cosine-comparable to projected user feature vectors in the interaction graph
    The entire retrieval mechanism in Eqs. 2 and 3 relies on this alignment. The linear layer only learns user-side projections; the query embeddings are frozen.
  • domain assumption Users who interacted with the ground-truth item are a suitable positive-training set for the subgraph retriever
    The training loss L1 (Eq. 4) treats neighbors of the ground-truth item as true users to align query embeddings. This assumes such users are representative of the correct retrieval target.
  • standard math Standard GNN message passing on the final subgraph captures high-order collaborative filtering information for down-stream ranking
    The GNN (GCN or GraphTransformer) is applied as in prior works; the paper relies on the established effectiveness of GNNs in recommendation (Section 3.5).

how reviews work

0 comments
Cite this review

Pith. "Pith review of CORONA: A Coarse-to-Fine Framework for Graph-based Recommendation with Large Language Models." pith.science (2026). https://pith.science/paper/RWDXVAKJ

@misc{pith2026250617281,
  author       = {Pith},
  title        = {Pith review of: CORONA: A Coarse-to-Fine Framework for Graph-based Recommendation with Large Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RWDXVAKJ}},
  note         = {Machine review of arXiv:2506.17281}
}
read the original abstract

Recommender systems (RSs) are designed to retrieve candidate items a user might be interested in from a large pool. A common approach is using graph neural networks (GNNs) to capture high-order interaction relationships. As large language models (LLMs) have shown strong capabilities across domains, researchers are exploring their use to enhance recommendation. However, prior work limits LLMs to re-ranking results or dataset augmentation, failing to utilize their power during candidate filtering - which may lead to suboptimal performance. Instead, we propose to leverage LLMs' reasoning abilities during the candidate filtering process, and introduce Chain Of Retrieval ON grAphs (CORONA) to progressively narrow down the range of candidate items on interaction graphs with the help of LLMs: (1) First, LLM performs preference reasoning based on user profiles, with the response serving as a query to extract relevant users and items from the interaction graph as preference-assisted retrieval; (2) Then, using the information retrieved in the previous step along with the purchase history of target user, LLM conducts intent reasoning to help refine an even smaller interaction subgraph as intent-assisted retrieval; (3) Finally, we employ a GNN to capture high-order collaborative filtering information from the extracted subgraph, performing GNN-enhanced retrieval to generate the final recommendation results. The proposed framework leverages the reasoning capabilities of LLMs during the retrieval process, while seamlessly integrating GNNs to enhance overall recommendation performance. Extensive experiments on various datasets and settings demonstrate that our proposed CORONA achieves state-of-the-art performance with an 18.6% relative improvement in recall and an 18.4% relative improvement in NDCG on average.

Figures

Figures reproduced from arXiv: 2506.17281 by the authors.

Figure 1
Figure 1. Comparisons between previous methods and our [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The overall framework of our proposed CORONA , which includes three stages of retrieval from different granularities. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Applying different LLMs in CORONA on three [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Hyperparameter experiments on three datasets in [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Total time cost (in seconds) of different LLM-based Intent Reasoning: The users viewing preferences lean toward seriousthematicand genre-specific filmsparticularly historical [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

78 extracted references · 49 canonical work pages

  1. [1]

    Keqin Bao, Jizhi Zhang, Yang Zhang, Wenjie Wang, Fuli Feng, and Xiangnan He

  2. [2]

    James Bennett, Stan Lanning, et al. 2007. The netflix prize. InIn Proceedings of KDD Cup and Workshop, Vol. 2007. New York, 35

    work page 2007

  3. [3]

    Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. 2020. Language models are few-shot learners.Advances in Neural Information Processing Systems33 (2020), 1877–1901

    2020

  4. [4]

    Lei Chen, Le Wu, Richang Hong, Kun Zhang, and Meng Wang. 2020. Revisiting graph based collaborative filtering: A linear residual graph convolutional network approach. InIn Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 34. 27–34

    work page 2020

  5. [5]

    Yuxin Chen, Junfei Tan, An Zhang, Zhengyi Yang, Leheng Sheng, Enzhi Zhang, Xiang Wang, and Tat-Seng Chua. 2024. On Softmax Direct Preference Optimiza- tion for Recommendation.arXiv preprint arXiv:2406.09215(2024)

    arXiv 2024

  6. [6]

    Wei-Lin Chiang, Zhuohan Li, Zi Lin, Ying Sheng, Zhanghao Wu, Hao Zhang, Lianmin Zheng, Siyuan Zhuang, Yonghao Zhuang, Joseph E Gonzalez, et al. 2023. Vicuna: An open-source chatbot impressing gpt-4 with 90%* chatgpt quality.See https://vicuna. lmsys. org (accessed 14 April 2023)2, 3 (2023), 6

    work page 2023

  7. [7]

    Sunhao Dai, Ninglu Shao, Haiyuan Zhao, Weijie Yu, Zihua Si, Chen Xu, Zhongx- iang Sun, Xiao Zhang, and Jun Xu. 2023. Uncovering chatgpt’s capabilities in recommender systems. InIn Proceedings of the 17th ACM Conference on Recom- mender Systems. 1126–1132

    work page 2023

  8. [8]

    Abhinandan S Das, Mayur Datar, Ashutosh Garg, and Shyam Rajaram. 2007. Google news personalization: scalable online collaborative filtering. InIn Pro- ceedings of the ACM Web Conference 2007. 271–280

    work page 2007

Show all 78 references
  1. [9]

    James Davidson, Benjamin Liebald, Junning Liu, Palash Nandy, Taylor Van Vleet, Ullas Gargi, Sujoy Gupta, Yu He, Mike Lambert, Blake Livingston, et al . 2010. The YouTube video recommendation system. InIn Proceedings of the 4th ACM conference on Recommender systems. 293–296

    2010

  2. [10]

    Chen Gao, Yu Zheng, Nian Li, Yinfeng Li, Yingrong Qin, Jinghua Piao, Yuhan Quan, Jianxin Chang, Depeng Jin, Xiangnan He, et al. 2023. A survey of graph neural networks for recommender systems: Challenges, methods, and directions. ACM Transactions on Recommender Systems1, 1 (20...

    2023

  3. [11]

    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)

    2023 arXiv

  4. [12]

    Binzong Geng, Zhaoxin Huan, Xiaolu Zhang, Yong He, Liang Zhang, Fajie Yuan, Jun Zhou, and Linjian Mo. 2024. Breaking the length barrier: Llm-enhanced CTR prediction in long textual user behaviors. InIn Proceedings of the 47th International ACM SIGIR Conference on Research and ...

    2024

  5. [13]

    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). InIn Proceedings of the 16th ACM Conference on Recommender Systems. 299–315

    2022

  6. [14]

    F Maxwell Harper and Joseph A Konstan. 2015. The movielens datasets: History and context.Acm Transactions on Interactive Intelligent Systems5, 4 (2015), 1–19

    2015

  7. [15]

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

    2020

  8. [16]

    Xiaoxin He, Yijun Tian, Yifei Sun, Nitesh Chawla, Thomas Laurent, Yann LeCun, Xavier Bresson, and Bryan Hooi. 2024. G-retriever: Retrieval-augmented gen- eration for textual graph understanding and question answering.Advances in Neural Information Processing Systems37 (2024), ...

    2024

  9. [17]

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

    2024

  10. [18]

    Wenyue Hua, Lei Li, Shuyuan Xu, Li Chen, and Yongfeng Zhang. 2023. Tutorial on large language models for recommendation. InIn Proceedings of the 17th ACM Conference on Recommender Systems. 1281–1283

    2023

  11. [19]

    Feiran Huang, Zefan Wang, Xiao Huang, Yufeng Qian, Zhetao Li, and Hao Chen

  12. [20]

    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)

    2024 arXiv

  13. [21]

    InIn Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval

    Aligning distillation for cold-start item recommendation. InIn Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1147–1157

  14. [22]

    Thomas N Kipf and Max Welling. 2016. Semi-supervised classification with graph convolutional networks.arXiv preprint arXiv:1609.02907(2016)

    2016 arXiv

  15. [23]

    Sein Kim, Hongseok Kang, Seungyoon Choi, Donghyun Kim, Minchul Yang, and Chanyoung Park. 2024. Large language models meet collaborative filtering: An efficient all-round llm-based recommender system. InIn Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and...

    2024

  16. [24]

    Yehuda Koren, Steffen Rendle, and Robert Bell. 2021. Advances in collaborative filtering.Recommender Systems Handbook(2021), 91–142

    2021

  17. [25]

    Xiaoyu Kong, Jiancan Wu, An Zhang, Leheng Sheng, Hui Lin, Xiang Wang, and Xiangnan He. 2024. Customizing Language Models with Instance-wise LoRA for Sequential Recommendation.arXiv preprint arXiv:2408.10159(2024)

    2024 arXiv

  18. [26]

    Jiayi Liao, Sihang Li, Zhengyi Yang, Jiancan Wu, Yancheng Yuan, Xiang Wang, and Xiangnan He. 2024. Llara: Large language-recommendation assistant. In In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1785–1795

    2024

  19. [27]

    Haoyang Li, Xin Wang, Ziwei Zhang, Jianxin Ma, Peng Cui, and Wenwu Zhu. 2021. Intention-aware sequential recommendation with structured intent transition. IEEE Transactions on Knowledge and Data Engineering34, 11 (2021), 5403–5414

    2021

  20. [28]

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

    2024

  21. [29]

    Jianghao Lin, Xinyi Dai, Yunjia Xi, Weiwen Liu, Bo Chen, Hao Zhang, Yong Liu, Chuhan Wu, Xiangyang Li, Chenxu Zhu, et al . 2025. How can recommender systems benefit from large language models: A survey.ACM Transactions on Information Systems43, 2 (2025), 1–47

    2025

  22. [30]

    Qidong Liu, Xian Wu, Xiangyu Zhao, Yejing Wang, Zijian Zhang, Feng Tian, and Yefeng Zheng. 2024. Large Language Models Enhanced Sequential Recommenda- tion for Long-tail User and Item.arXiv preprint arXiv:2405.20646(2024)

    2024 arXiv

  23. [31]

    Peng Liu, Lemei Zhang, and Jon Atle Gulla. 2023. Pre-train, Prompt, and Recom- mendation: A Comprehensive Survey of Language Modeling Paradigm Adapta- tions in Recommender Systems.Transactions of the Association for Computational Linguistics11 (2023), 1553–1571

    2023

  24. [32]

    Qiyao Ma, Xubin Ren, and Chao Huang. 2024. XRec: Large Language Models for Explainable Recommendation.arXiv preprint arXiv:2406.02377(2024)

    2024 arXiv

  25. [33]

    Weiwen Liu, Qing Liu, Ruiming Tang, Junyang Chen, Xiuqiang He, and Pheng Ann Heng. 2020. Personalized Re-ranking with Item Relationships for E-commerce. InIn Proceedings of the 29th ACM International Conference on Infor- mation & Knowledge Management. 925–934

    2020

  26. [34]

    Jianmo Ni, Jiacheng Li, and Julian McAuley. 2019. Justifying recommendations using distantly-labeled reviews and fine-grained aspects. InIn Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natu...

    2019

  27. [35]

    Sheshera Mysore, Andrew McCallum, and Hamed Zamani. 2023. Large language model augmented narrative driven recommendations. InIn Proceedings of the 17th ACM Conference on Recommender Systems. 777–783

    2023

  28. [36]

    N Reimers. 2019. Sentence-BERT: Sentence Embeddings using Siamese BERT- Networks.arXiv preprint arXiv:1908.10084(2019)

    2019 arXiv

  29. [37]

    Fabio Petroni, Tim Rocktäschel, Patrick Lewis, Anton Bakhtin, Yuxiang Wu, Alexander H Miller, and Sebastian Riedel. 2019. Language models as knowledge bases?arXiv preprint arXiv:1909.01066(2019)

    2019 arXiv

  30. [38]

    Yankun Ren, Zhongde Chen, Xinxing Yang, Longfei Li, Cong Jiang, Lei Cheng, Bo Zhang, Linjian Mo, and Jun Zhou. 2024. Enhancing sequential recommenders with augmented knowledge from aligned large language models. InIn Proceedings of the 47th International ACM SIGIR Conference o...

    2024

  31. [39]

    Xubin Ren, Wei Wei, Lianghao Xia, Lixin Su, Suqi Cheng, Junfeng Wang, Dawei Yin, and Chao Huang. 2024. Representation learning with large language models for recommendation. InIn Proceedings of the ACM Web Conference 2024. 3464– 3475

    2024

  32. [40]

    Wentao Shi, Xiangnan He, Yang Zhang, Chongming Gao, Xinyue Li, Jizhi Zhang, Qifan Wang, and Fuli Feng. 2024. Enhancing Long-Term Recommendation with Bi- level Learnable Large Language Model Planning.arXiv preprint arXiv:2403.00843 (2024)

    2024 arXiv

  33. [41]

    Scott Sanner, Krisztian Balog, Filip Radlinski, Ben Wedin, and Lucas Dixon

  34. [42]

    InIn Proceedings of the 17th ACM Conference on Recommender Systems

    Large language models are competitive near cold-start recommenders for language-and item-based preferences. InIn Proceedings of the 17th ACM Conference on Recommender Systems. 890–896

  35. [43]

    Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al. 2023. Llama: Open and efficient foundation language models.arXiv preprint arXiv:2302.13971(2023)

    2023 arXiv

  36. [44]

    Zhu Sun, Hongyang Liu, Xinghua Qu, Kaidong Feng, Yan Wang, and Yew Soon Ong. 2024. Large language models for intent-driven session recommendations. InIn Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval. 324–334

    2024

  37. [45]

    Juntao Tan, Shuyuan Xu, Wenyue Hua, Yingqiang Ge, Zelong Li, and Yongfeng Zhang. 2024. Towards llm-recsys alignment with textual id learning.arXiv preprint arXiv:2403.19021(2024)

    2024 arXiv

  38. [46]

    Jiayin Wang, Fengran Mo, Weizhi Ma, Peijie Sun, Min Zhang, and Jian-Yun Nie

  39. [47]

    Maksims Volkovs, Guangwei Yu, and Tomi Poutanen. 2017. Dropoutnet: Ad- dressing cold start in recommender systems.Advances in Neural Information Processing Systems30 (2017). CORONA: A Coarse-to-Fine Framework for Graph-based Recommendation with Large Language Models SIGIR ’25,...

    2017

  40. [48]

    Jie Wang, Alexandros Karatzoglou, Ioannis Arapakis, and Joemon M Jose. 2024. Reinforcement Learning-based Recommender Systems with Large Language Models for State Reward and Action Modeling. InIn Proceedings of the 47th International ACM SIGIR Conference on Research and Develo...

    2024

  41. [49]

    Y. Wang, Z. Chu, X. Ouyang, S. Wang, H. Hao, Y. Shen, J. Gu, S. Xue, J. Zhang, Q. Cui, L. Li, J. Zhou, and S. Li. 2024. LLMRG: Improving Recommendations through Large Language Model Reasoning Graphs. InIn Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 38. ...

    2024

  42. [50]

    Yifan Wang, Suyao Tang, Yuntong Lei, Weiping Song, Sheng Wang, and Ming Zhang. 2020. Disenhan: Disentangled heterogeneous graph attention network for recommendation. InIn Proceedings of the 29th ACM International Conference on Information & Knowledge Management. 1605–1614

    2020

  43. [51]

    Xiang Wang, Xiangnan He, Meng Wang, Fuli Feng, and Tat-Seng Chua. 2019. Neural graph collaborative filtering. InIn Proceedings of the 42nd international ACM SIGIR conference on Research and development in Information Retrieval. 165–174

    2019

  44. [52]

    Xiaolei Wang, Xinyu Tang, Wayne Xin Zhao, Jingyuan Wang, and Ji-Rong Wen

  45. [53]

    Rethinking the evaluation for conversational recommendation in the era of large language models.arXiv preprint arXiv:2305.13112(2023)

    2023 arXiv

  46. [54]

    Jiajing Xu, Andrew Zhai, and Charles Rosenberg. 2022. Rethinking personalized ranking at Pinterest: An end-to-end approach. InIn Proceedings of the 16th ACM Conference on Recommender Systems. 502–505

    2022

  47. [55]

    Shenghao Yang, Weizhi Ma, Peijie Sun, Qingyao Ai, Yiqun Liu, Mingchen Cai, and Min Zhang. 2024. Sequential recommendation with latent relations based on large language model. InIn Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Informa...

    2024

  48. [56]

    Wei Wei, Xubin Ren, Jiabin Tang, Qinyong Wang, Lixin Su, Suqi Cheng, Junfeng Wang, Dawei Yin, and Chao Huang. 2024. Llmrec: Large language models with graph augmentation for recommendation. InIn Proceedings of the 17th ACM International Conference on Web Search and Data Mining...

    2024

  49. [57]

    Yinwei Wei, Xiang Wang, Liqiang Nie, Xiangnan He, and Tat-Seng Chua. 2020. Graph-refined convolutional network for multimedia recommendation with implicit feedback. InIn Proceedings of the 28th ACM International Conference on Multimedia. 3541–3549

    2020

  50. [58]

    Likang Wu, Zhi Zheng, Zhaopeng Qiu, Hao Wang, Hongchao Gu, Tingjia Shen, Chuan Qin, Chen Zhu, Hengshu Zhu, Qi Liu, et al . 2024. A survey on large language models for recommendation.World Wide Web27, 5 (2024), 60

    2024

  51. [59]

    Jizhi Zhang, Keqin Bao, Yang Zhang, Wenjie Wang, Fuli Feng, and Xiangnan He. 2023. Is chatgpt fair for recommendation? evaluating fairness in large lan- guage model recommendation. InIn Proceedings of the 17th ACM Conference on Recommender Systems. 993–999

    2023

  52. [60]

    Junjie Zhang, Ruobing Xie, Yupeng Hou, Xin Zhao, Leyu Lin, and Ji-Rong Wen

  53. [61]

    Yuhao Yang, Lianghao Xia, Da Luo, Kangyi Lin, and Chao Huang. 2024. GraphPro: Graph Pre-training and Prompt Learning for Recommendation. InIn Proceedings of the ACM Web Conference 2024. 3690–3699

    2024

  54. [62]

    Rex Ying, Ruining He, Kaifeng Chen, Pong Eksombatchai, William L Hamilton, and Jure Leskovec. 2018. Graph convolutional neural networks for web-scale recommender systems. InIn Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 974–983

    2018

  55. [63]

    Seongjun Yun, Minbyul Jeong, Raehyun Kim, Jaewoo Kang, and Hyunwoo J Kim

  56. [64]

    Yang Zhang, Fuli Feng, Jizhi Zhang, Keqin Bao, Qifan Wang, and Xiangnan He

  57. [65]

    Wayne Xin Zhao, Kun Zhou, Junyi Li, Tianyi Tang, Xiaolei Wang, Yupeng Hou, Yingqian Min, Beichen Zhang, Junjie Zhang, Zican Dong, et al. 2023. A survey of large language models.arXiv preprint arXiv:2303.18223(2023)

    2023 arXiv

  58. [66]

    Yurou Zhao, Yiding Sun, Ruidong Han, Fei Jiang, Lu Guan, Xiang Li, Wei Lin, Weizhi Ma, and Jiaxin Mao. 2024. Aligning Explanations for Recommendation with Rating and Feature via Maximizing Mutual Information. InIn Proceedings of the 33rd ACM International Conference on Informa...

    2024

  59. [67]

    Recommendation as instruction following: A large language model em- powered recommendation approach.ACM Transactions on Information Systems (2023)

    2023

  60. [68]

    Kaike Zhang, Qi Cao, Yunfan Wu, Fei Sun, Huawei Shen, and Xueqi Cheng. 2024. Lorec: Large language model for robust sequential recommendation against poisoning attacks.arXiv preprint arXiv:2401.17723(2024)

    2024 arXiv

  61. [69]

    Xiaoyu Zhang, Yishan Li, Jiayin Wang, Bowen Sun, Weizhi Ma, Peijie Sun, and Min Zhang. 2024. Large language models as evaluators for recommendation explanations. InIn Proceedings of the 18th ACM Conference on Recommender Systems. 33–42

    2024

  62. [70]

    Yang Zhang, Keqin Bao, Ming Yan, Wenjie Wang, Fuli Feng, and Xiangnan He

  63. [71]

    Text-like Encoding of Collaborative Information in Large Language Models for Recommendation.arXiv preprint arXiv:2406.03210(2024)

    2024 arXiv

  64. [76]

    Zihuai Zhao, Wenqi Fan, Jiatong Li, Yunqing Liu, Xiaowei Mei, Yiqi Wang, Zhen Wen, Fei Wang, Xiangyu Zhao, Jiliang Tang, et al. 2024. Recommender systems in the era of large language models (llms).IEEE Transactions on Knowledge and Data Engineering(2024)

    2024

  65. [77]

    Yu Zheng, Chen Gao, Xiangnan He, Yong Li, and Depeng Jin. 2020. Price-aware recommendation with graph convolutional networks. In2020 IEEE 36th Interna- tional Conference on Data Engineering. IEEE, 133–144

    2020

  66. [78]

    Yaochen Zhu, Liang Wu, Qi Guo, Liangjie Hong, and Jundong Li. 2024. Collabo- rative large language model for recommender systems. InIn Proceedings of the ACM Web Conference 2024. 3162–3172

    2024

  67. [2019]

    Graph transformer networks.Advances in Neural Information Processing Systems32 (2019)

    2019

  68. [2023]

    InIn Proceedings of the 17th ACM Conference on Recommender Systems

    Tallrec: An effective and efficient tuning framework to align large language model with recommendation. InIn Proceedings of the 17th ACM Conference on Recommender Systems. 1007–1014

  69. [2024]

    InIn Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing

    A User-Centric Multi-Intent Benchmark for Evaluating Large Language Models. InIn Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 3588–3612

    2024

  70. [2025]

    Collm: Integrating collaborative embeddings into large language models for recommendation.IEEE Transactions on Knowledge and Data Engineering(2025)

    2025

Pith tools

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