Intent Representation Learning with Large Language Model for Recommendation

Lei Sang; Yiwen Zhang; Yi Zhang; Yu Wang

REVIEW 4 major objections 5 minor 1 cited by

Intent Representation Learning with Large Language Model for Recommendation

T0 review · 4 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read Aligning LLM-written intent summaries with interaction intents consistently improves graph-based recommenders across three datasets.

desk verdict Competently engineered intent-alignment framework with credible but not fully verified gains; the leakage risk and the altered AlphaRec baseline need to be settled before the empirical claim is accepted. read the letter →

arxiv 2502.03307 v4 pith:E2LY437A submitted 2025-02-05 cs.IR

Yu Wang , Lei Sang , Yi Zhang , Yiwen Zhang This is my paper · ORCID

classification cs.IR

keywords intentrepresentationlearninglargelanguagemodelsforrecommendationmultimodalalignmentmomentumdistillationcollaborativefilteringcontrastivetext-enhancedgraphneuralnetworks

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

The paper sets out to show that recommenders get better when the latent reasons behind user-item interactions, called intents, are modelled in two modalities at once. A large language model reads user reviews and item descriptions to write a short intent summary of what the user likes and dislikes, while a graph encoder learns interaction intents from the collaborative signal. IRLLRec is the model-agnostic wrapper that aligns these two representation spaces and matches them, and the paper reports consistent Recall and NDCG gains over strong LLM-enhanced baselines on Amazon-book, Yelp, and Amazon-movie. The appeal is practical: if the claim is right, any graph-based recommender can be improved by plugging in intent alignment without redesigning the base model.

What carries the argument

The central object is the multimodal intent pair $\mathbf{z}$ and $\mathbf{r}$: a textual intent embedding $\mathbf{z}$ produced by an LLM summary and a linear text encoder, and an interaction intent embedding $\mathbf{r}$ obtained by weighting $K$ intent prototypes with softmax attention over graph embeddings. Two mechanisms do the work: Intent Alignment, which combines a pairwise contrastive loss with translation alignment that adds Gaussian noise to each modality and treats the diagonal of the similarity matrix as positive pairs, and Interaction-text Matching, which uses momentum distillation with an EMA-updated teacher encoder to generate pseudo-labels for student matching scores. This teacher-student setup is what lets the model pick out the latent key intents among noisy text and interactions.

What would settle it

Inspect the released code or rerun the pipeline with a temporal split (train on early reviews, test on later ones) and with user prompts built only from training-set reviews; if the Recall and NDCG advantages over RLMRec disappear or shrink sharply, the central superiority claim was driven by test information leaking into the LLM summaries.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that intent is not a single hidden variable but a pair of complementary views that must be aligned and matched. Textual intents are produced by chain-of-thought prompts that aggregate the positive and negative attributes of every item a user interacted with, while interaction intents are computed as an expectation over K learned intent prototypes on LightGCN embeddings. The paper's contribution is a dual-tower framework in which pairwise InfoNCE alignment and Gaussian-noise translation alignment bring the two spaces together, and Interaction-text Matching with momentum distillation learns which pieces of text correspond to which interaction intents. Across five base models and three datasets, the framework reports statistically significant improvements over RLMRec, KAR, and AlphaRec, with the largest gains appearing in the sparsest user groups.

Load-bearing premise

The evaluation assumes that the LLM intent summaries are generated from training interactions only, so that reviews assigned to the test split never influence the prompts; if they do, the reported improvements could reflect leakage rather than generalization.

Editorial extensions

If this is right

Any graph-based collaborative filtering model can be wrapped with IRLLRec's intent alignment and matching modules and should show higher Recall@20 and NDCG@20 on review-rich datasets.
Sparse and cold-start users benefit most, because the LLM's textual intent carries preference signal before enough interactions accumulate.
The framework makes recommendations more interpretable: the aligned textual intent is a readable natural-language statement of why a user or item behaves as it does.
Better text-embedding models directly translate into better recommendations, since the framework is bottlenecked by the quality of the LLM intent representation.
Removing either the alignment or the matching module degrades performance below the base model, so the two losses are jointly load-bearing.

Reading between the lines

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

A temporal split or a prompt-construction audit would settle whether the reported gains come from genuine intent alignment or from test reviews leaking into the LLM summaries through the aggregation in Eq. (4).
The same dual-tower alignment recipe could be applied to other side information, such as images or audio, where a second modality carries preference signals not visible in the interaction graph.
Because the matching module is borrowed from vision-language pretraining, the framework suggests that recommendation can borrow more cross-modal matching machinery from that literature.
Formalizing the expectation in Eqs. (5)-(6) may yield a variational bound that predicts when intent prototypes help, turning the empirical recipe into a testable model-selection rule.

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.

Desk Editor's Note

read the letter

Two things to know before you spend time on this. The architecture is a well-built combination of known blocks: CoT-summarized textual intents, learnable intent prototypes, InfoNCE-style alignment, and momentum distillation for text-interaction matching. That specific composition is new, and the paper does several things well: it releases code, tests across five backbones on three datasets, includes ablations and hyperparameter sensitivity, and the gains (roughly 1-15% over LLM-enhanced baselines) are believable in magnitude. The cold-start analysis is a nice touch. This deserves a serious referee. The soft spots are real but addressable. The biggest is potential leakage. User intents are defined as an aggregate over the user's interaction set (Eq. 4), and the prompt in Fig. 8 explicitly includes each purchased item's review text. Section 4.1.1 only mentions k-core filtering and random splits; it never states that validation/test interactions are excluded from the LLM prompt. If prompts are built before splitting, the user embedding can directly encode test item identities, and the cold-start gains in Fig. 3 would be inflated. This is checkable with the released code, but the paper needs to state it clearly and, if the protocol does leak, rerun. Without that, the central empirical claim is not fully supported. Second, the AlphaRec baseline is altered. The paper says they used RLMRec's profile representation as the item embedding instead of AlphaRec's proposed item title, "for fairness." That is not the published AlphaRec, and it likely handicaps it. The comparison should be redone with the original setup, or the justification strengthened substantially. Third, the significance claims are overstated. Table 2 has several cells without stars (e.g., SGL on Yelp: Recall@5, NDCG@5, NDCG@20), yet the text claims "significant improvements across all metrics" and mentions only one exception. Also no standard deviations or test details are given, so the pairwise significance tests are hard to verify. Minor: the time complexity analysis writes O((M+N)^2 d) for the loss, which should depend on batch size, not the full user/item count. Sloppy but not fatal. The math in Eqs. 5-6 is motivational rather than operational. Overall: the paper is interesting and the experiments are extensive, but the leakage question and the altered baseline need to be resolved before the superiority claim is accepted. I would send it to peer review with a request to address these points. If the leakage is confirmed, major revision; if not, it's a solid contribution.

Referee Report

4 major / 5 minor

Summary. The paper proposes IRLLRec, a model-agnostic framework that combines LLM-constructed textual intents with interaction-based intents for collaborative filtering. The method uses a dual-tower architecture with a text encoder and a graph encoder, pairwise and translation alignment losses to bring the two intent spaces together, and an interaction-text matching module based on momentum distillation. Experiments on Amazon-book, Yelp, and Amazon-movie report consistent Recall and NDCG improvements over KAR, RLMRec, AlphaRec, and several base models, with claimed statistical significance.

Significance. If the empirical findings are valid, IRLLRec would be a useful contribution to intent-based and LLM-enhanced recommendation, showing that fine-grained textual intents can be aligned with interaction intents in a plug-in manner. The paper provides a public code link and uses standard public datasets. The main value is the proposed combination of existing techniques (contrastive alignment, translation alignment, momentum distillation) in a model-agnostic package; the individual components are not entirely new, but their integration is clearly presented. However, the significance hinges on the evaluation being free of leakage and on the statistical claims being accurate.

major comments (4)

Potential test-set leakage: Eq. (4) defines the textual user intent I_u as the aggregate over all items in I_u, and Figure 8 shows that the LLM prompt includes each purchased item's title and review text. The dataset split described in Section 4.1.1 is purely random (3:1:1 or 8:1:1), and the paper never states that validation/test interactions are excluded from the prompt construction. If the prompts are generated from the full interaction set before splitting, then for a typical user the textual intent embedding contains titles and reviews of items that later appear in the test set. Since IRLLRec aligns textual and interaction intents via contrastive loss and scores recommendations by the inner product of user and item representations (Eq. 1), this could directly leak test item identities into the user embedding. The reported gains, especially for the low-interaction groups in Figure 3, could then reflect leakage rather than true intent alignment. The authors must either confirm explicitly (in text or code) that LLM prompts use only training interactions, or re-run the experiments with split-aware intent construction and report the results. This is load-bearing for the central empirical claim.
[Table 2, Section 4.2.1] The text claims that IRLLRec shows 'significant improvements across all metrics', but Table 2 contradicts this. In the SGL block on Yelp, IRLLRec has no asterisk for Recall@5 (0.0465) and NDCG@20 (0.0805), and the Recall@5 improvement is reported as -0.43% relative to RLMRec-Gen. Moreover, several other cells in the same block lack asterisks. The significance claim should be corrected to match the table, and the authors should explain which comparisons are significant and which are not.
[Section 4.2.1] The authors state that AlphaRec's poor performance 'may stem from our adherence to fairness, where we use RLMRec's profile representation as the item embedding instead of the proposed item title.' This means the reported AlphaRec numbers are not produced by the original AlphaRec method, but by a modified variant. Comparing against a modified, possibly suboptimal version of a baseline weakens the claim that IRLLRec outperforms AlphaRec. The authors should either run the original AlphaRec implementation or clearly label this as a variant and justify why the modification is necessary.
[Section 4.2.1, Table 2] No measure of variance is reported. The paper states that results are averaged over 5 runs and places asterisks for statistical significance, but no standard deviations, confidence intervals, or details of the significance test (e.g., paired t-test or Wilcoxon, what is paired, across which runs) are provided. Without this information, the significance asterisks cannot be independently verified. Please report standard deviations, at least for the main results in Table 2, and describe the test procedure in Section 4.1.3.

minor comments (5)

In the definition of the translation alignment perturbation, the second noise variable is typeset as 𝝐𝑟 instead of 𝝐𝑧, which is confusing since both modalities are perturbed independently.
The individual losses L_user_pair, L_pos_pair, and L_neg_pair are introduced but never explicitly defined. Please specify how each is computed from Eq. (13), or clarify that they are obtained by applying the same InfoNCE loss to different sample types in the batch.
The loss L_ITM uses 'L_tran' without a definition; presumably it is the sum of L_text_tran and L_inter_tran from Eq. (15), but this should be stated explicitly.
The phrase 'cold-start (not strictly defined)' is informal for a published paper. Since the sparsity analysis is used to support a specific claim about cold-start users, please define the user groups precisely (e.g., by interaction count ranges) rather than relying on an informal label.
The paper evaluates five base models but does not describe how IRLLRec modifies each base model's training objective beyond adding the auxiliary losses. A short paragraph stating whether the base model's original loss is retained unchanged, and how the intent alignment loss is combined with it, would improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation found; the claimed gains are empirical and the self-citations are not load-bearing.

full rationale

No circular derivation chain was found. IRLLRec's central claim is empirical superiority on Recall and NDCG, which is measured directly against baselines on held-out splits rather than derived from its own inputs. The LLM intent construction in Eq. 4 is an input feature produced from reviews and descriptions; the intent prototypes in Eqs. 9-10 are learned parameters trained with the auxiliary losses in Eqs. 13-17, whose objectives (mutual-information-style alignment and similarity matching) do not contain the evaluation metrics Recall@k or NDCG@k. The statistical identities in Eqs. 5-6 are standard marginalization and expectation-approximation steps, cited to Refs. [31, 39], and they do not define the target metric. The graph-structure-learning motivation cites the authors' prior work [34], and BIGCF as a base model cites [51], but the mechanisms used in this paper are fully specified in the paper itself (e.g., Eq. 11) and are not used to forbid alternative designs or to import an unverified uniqueness theorem; these self-citations are therefore not load-bearing. The main risk visible in the manuscript is potential evaluation leakage if test-set reviews enter the LLM prompts of Eq. 4 before the random split described in Section 4.1.1; that is an evaluation-validity and correctness concern, not a circular-derivation concern, and would be scored under correctness risk rather than circularity.

Assumptions & free parameters 3 free parameters · 4 assumptions · 1 invented entities

The empirical claim rests on tuning multiple hyperparameters and on LLM-generated text features; no closed-form derivation links model inputs to the reported metrics. The main uncharged cost is the assumption that LLM summaries do not leak test-period information and that the hand-set prototype count and loss weights are adequate.

free parameters (3)

Number of intent prototypes K = not stated
K defines the discrete intent inventory in Eq. 9-10 and is chosen by hand; the paper never reports the value used, which affects reproducibility.
Loss weights lambda1, lambda2, lambda3 = not stated
Tuned on validation data (Figure 6); only ranges are reported, not the exact values used for the main results in Table 2.
Contrastive temperature tau, scaling factor eta, ITM balance alpha=0.4, momentum beta=0.999 = partial: alpha and beta stated; tau and eta not stated
These hand-set hyperparameters control the strength of alignment and teacher-student matching; missing tau and eta values make exact reproduction harder.

assumptions (4)

domain assumption Factorization and linear-exchange approximation in Eq. 5-6
The paper assumes P(y|u,i)=E[P(y|c_u,c_i)] and approximates E[f(c)] by f(E[c]); this is cited to [31,39] but not derived, and the required linearity of f is not justified.
domain assumption LLM CoT summaries faithfully capture true user/item intents
The text tower is only useful if GPT-4o-mini outputs accurate, stable intent summaries; hallucinated or generic summaries would break the alignment and degrade the method.
domain assumption No leakage from test reviews into LLM prompts
Section 4.1.1 describes a random split with no stated exclusion of test reviews from Eq. 4; the paper implicitly assumes text intent construction is independent of test labels.
domain assumption Graph structure learning recovers a clean intent graph
Eq. 11-12 assume the learned mask M removes noise without removing true preference signal, which is a modeling premise inherited from GSL literature.

invented entities (1)

Learnable intent prototype vectors c_u^k and c_i^k
purpose: Map user and item embeddings to intent-aware interaction representations in Eq. 9-10
K prototype vectors are trained with the recommendation loss and have no falsifiable prediction outside the model; they are internal latent variables introduced by the method.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Intent Representation Learning with Large Language Model for Recommendation." pith.science (2026). https://pith.science/paper/E2LY437A

@misc{pith2026250203307,
  author       = {Pith},
  title        = {Pith review of: Intent Representation Learning with Large Language Model for Recommendation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/E2LY437A}},
  note         = {Machine review of arXiv:2502.03307}
}

read the original abstract

Intent-based recommender systems have garnered significant attention for uncovering latent fine-grained preferences. Intents, as underlying factors of interactions, are crucial for improving recommendation interpretability. Most methods define intents as learnable parameters updated alongside interactions. However, existing frameworks often overlook textual information (e.g., user reviews, item descriptions), which is crucial for alleviating the sparsity of interaction intents. Exploring these multimodal intents, especially the inherent differences in representation spaces, poses two key challenges: i) How to align multimodal intents and effectively mitigate noise issues; ii) How to extract and match latent key intents across modalities. To tackle these challenges, we propose a model-agnostic framework, Intent Representation Learning with Large Language Model (IRLLRec), which leverages large language models (LLMs) to construct multimodal intents and enhance recommendations. Specifically, IRLLRec employs a dual-tower architecture to learn multimodal intent representations. Next, we propose pairwise and translation alignment to eliminate inter-modal differences and enhance robustness against noisy input features. Finally, to better match textual and interaction-based intents, we employ momentum distillation to perform teacher-student learning on fused intent representations. Empirical evaluations on three datasets show that our IRLLRec framework outperforms baselines.Code available at https://github.com/wangyu0627/IRLLRec.

Figures

Figures reproduced from arXiv: 2502.03307 by the authors.

**Figure 1.** (a) User-item interaction bipartite graph; (b) Disentangled interactions incorporating user intents: 𝑢1-𝑖3 is influenced by intents 𝑐2 and 𝑐𝑛, reflecting a preference for businesses offering fine dining, convenience, and leisure; (c) Gaussian kernel density estimation (KDE [38]) visualizes three embedding types: interaction from the pre-trained LightGCN [13], profile from RLMRec [29] extracted attribute summaries, a… view at source ↗

**Figure 2.** Illustration of IRLLRec. Multi Intent Fusion (MIF): [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Performance comparison of different sparsity levels. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Figures from the paper (5 more)

**Figure 4.** Figure 4: Ablation studies of model variants on the Amazon [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

**Figure 5.** Figure 5: Case study on Amazon movie. (c) The coefficient effect of ଷon Recall@20 (b) The coefficient effect of ଶon Recall@20 (a) The coefficient effect of ଵon Recall@20 [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

**Figure 6.** Figure 6: Performance comparison w.r.t. loss coefficients ap [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]

**Figure 7.** Figure 7: Case study on LLM-based item intent generation in [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]

**Figure 8.** Figure 8: Case study on LLM-based user intent generation in [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Heterogeneous Graph Masked Contrastive Learning for Robust Recommendation
cs.IR 2025-05 conditional novelty 4.0 of 10

MCL improves heterogeneous-graph recommendation accuracy and robustness by combining random masking and random propagation with cross-view contrastive learning on one-hop and meta-path neighborhoods.

Reference graph

Works this paper leans on

56 extracted references · 43 canonical work pages · cited by 1 Pith paper

[1]

Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Floren- cia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 Technical Report. arXiv preprint arXiv:2303.08774 (2023)

arXiv 2023
[2]

Keqin Bao, Jizhi Zhang, Yang Zhang, Wenjie Wang, Fuli Feng, and Xiangnan He. 2023. Tallrec: An Effective and Efficient Tuning Framework to Align lLarge Language Model with Recommendation. In RecSys. 1007–1014

work page 2023
[3]

Xuheng Cai, Chao Huang, Lianghao Xia, and Xubin Ren. 2023. LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation. In ICLR

work page 2023
[4]

Hong Chen, Yudong Chen, Xin Wang, Ruobing Xie, Rui Wang, Feng Xia, and Wenwu Zhu. 2021. Curriculum Disentangled Recommendation with Noisy Multi- feedback. NeurIPS 34 (2021), 26924–26936

work page 2021
[5]

Lei Chen, Le Wu, Richang Hong, Kun Zhang, and Meng Wang. 2020. Revisiting Graph based Collaborative Filtering: A Linear Residual Graph Convolutional Network Approach. In AAAI, Vol. 34. 27–34

work page 2020
[6]

Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. 2020. A Simple Framework for Contrastive Learning of Visual Representations. In ICML. PMLR, 1597–1607

work page 2020
[7]

Yongjun Chen, Zhiwei Liu, Jia Li, Julian McAuley, and Caiming Xiong. 2022. Intent Contrastive Learning for Sequential Recommendation. In WWW. 2172–2182

work page 2022
[8]

Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al. 2024. The Llama 3 Herd of Models. arXiv preprint arXiv:2407.21783 (2024)

arXiv 2024

Show all 56 references

[9]

Shaohua Fan, Junxiong Zhu, Xiaotian Han, Chuan Shi, Linmei Hu, Biyu Ma, and Yongliang Li. 2019. Metapath-guided Heterogeneous Graph Neural Network for Intent Recommendation. In KDD. 2478–2486

2019
[10]

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
[11]

Xavier Glorot and Yoshua Bengio. 2010. Understanding the Difficulty of Training Deep Feedforward Neural Networks. In Proceedings of the thirteenth International Conference on Artificial Intelligence and Statistics (ICAIS) . JMLR Workshop and Conference Proceedings, 249–256

2010
[12]

Kaiming He, Haoqi Fan, Yuxin Wu, Saining Xie, and Ross Girshick. 2020. Mo- mentum Contrast for Unsupervised Visual Representation Learning. In CVPR. 9729–9738

2020
[13]

Xiangnan He, Kuan Deng, Xiang Wang, Yan Li, Yongdong Zhang, and Meng Wang. 2020. Lightgcn: Simplifying and Powering Graph Convolution Network for Recommendation. In SIGIR. 639–648

2020
[14]

Xiangnan He, Lizi Liao, Hanwang Zhang, Liqiang Nie, Xia Hu, and Tat-Seng Chua. 2017. Neural Collaborative Filtering. In WWW. 173–182

2017
[15]

Jonathan Ho, Ajay Jain, and Pieter Abbeel. 2020. Denoising Diffusion Probabilistic Models. In NeurIPS, Vol. 33. 6840–6851

2020
[16]

Yupeng Hou, Zhankui He, Julian McAuley, and Wayne Xin Zhao. 2023. Learning Vector-quantized Item Representation for Transferable Sequential Recommenders. In WWW. 1162–1171

2023
[17]

Yupeng Hou, Jiacheng Li, Zhankui He, An Yan, Xiusi Chen, and Julian McAuley
[18]

Selvaraju, Akhilesh Deepak Gotmare, Shafiq Joty, Caiming Xiong, and Steven Hoi

Junnan Li, Ramprasaath R. Selvaraju, Akhilesh Deepak Gotmare, Shafiq Joty, Caiming Xiong, and Steven Hoi. 2021. Align before Fuse: Vision and Language Representation Learning with Momentum Distillation. In NeurIPS

2021
[19]

Dawen Liang, Rahul G Krishnan, Matthew D Hoffman, and Tony Jebara. 2018. Variational Autoencoders for Collaborative Filtering. In WWW. 689–698

2018
[20]

Jiayi Liao, Xiangnan He, Ruobing Xie, Jiancan Wu, Yancheng Yuan, Xingwu Sun, Zhanhui Kang, and Xiang Wang. 2024. RosePO: Aligning LLM-based Recom- menders with Human Values. arXiv preprint arXiv:2410.12519 (2024)

2024 arXiv
[21]

Jiayi Liao, Sihang Li, Zhengyi Yang, Jiancan Wu, Yancheng Yuan, Xiang Wang, and Xiangnan He. 2024. Llara: Large Language-Recommendation Assistant. In SIGIR. 1785–1795

2024
[22]

Dongsheng Luo, Wei Cheng, Wenchao Yu, Bo Zong, Jingchao Ni, Haifeng Chen, and Xiang Zhang. 2021. Learning to Drop: Robust Graph Neural Network via Topological Denoising. In WSDM. 779–787

2021
[23]

Jianxin Ma, Peng Cui, Kun Kuang, Xin Wang, and Wenwu Zhu. 2019. Disentangled Graph Convolutional Networks. In ICML. PMLR, 4212–4221

2019
[24]

Jianxin Ma, Chang Zhou, Peng Cui, Hongxia Yang, and Wenwu Zhu. 2019. Learn- ing Disentangled Representations for Recommendation. NeurIPS 32 (2019)

2019
[25]

Rui Meng, Ye Liu, Shafiq Rayhan Joty, Caiming Xiong, Yingbo Zhou, and Semih Yavuz. 2024. SFR-Embedding-Mistral: Enhance Text Retrieval with Transfer Learning. Salesforce AI Research Blog 3 (2024)

2024
[26]

Arvind Neelakantan, Tao Xu, Raul Puri, Alec Radford, Jesse Michael Han, Jerry Tworek, Qiming Yuan, Nikolas Tezak, Jong Wook Kim, Chris Hallacy, et al
[27]

Yingtao Peng, Chen Gao, Yu Zhang, Tangpeng Dan, Xiaoyi Du, Hengliang Luo, Yong Li, and Xiaofeng Meng. 2024. Denoising Alignment with Large Language Model for Recommendation. ACM Transactions on Information Systems (TOIS) (2024)

2024
[28]

Xubin Ren and Chao Huang. 2024. EasyRec: Simple yet Effective Language Models for Recommendation. arXiv preprint arXiv:2408.08821 (2024)

2024
[29]

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. In WWW. 3464–3475

2024
[30]

Xubin Ren, Lianghao Xia, Yuhao Yang, Wei Wei, Tianle Wang, Xuheng Cai, and Chao Huang. 2024. SSLRec: A Self-Supervised Learning Framework for Recommendation. In WSDM. 567–575

2024
[31]

Xubin Ren, Lianghao Xia, Jiashu Zhao, Dawei Yin, and Chao Huang. 2023. Dis- entangled Contrastive Collaborative Filtering. In SIGIR. 1137–1146

2023
[32]

Francesco Ricci, Lior Rokach, and Bracha Shapira. 2010. Introduction to Rec- ommender Systems Handbook. In Recommender Systems Handbook . Springer, 1–35

2010
[33]

Lei Sang, Yu Wang, Yiwen Zhang, and Xindong Wu. 2024. Denoising Heteroge- neous Graph Pre-training Framework for Recommendation. ACM Transactions on Information Systems (TOIS) (2024)

2024
[34]

Lei Sang, Yu Wang, Yi Zhang, Yiwen Zhang, and Xindong Wu. 2025. Intent- guided Heterogeneous Graph Contrastive Learning for Recommendation. IEEE Transactions on Knowledge and Data Engineering (TKDE) 37, 4 (2025), 1915–1929

2025
[35]

Lei Sang, Min Xu, Shengsheng Qian, Matt Martin, Peter Li, and Xindong Wu. 2020. Context-dependent Propagating-based Video Recommendation in Multimodal Heterogeneous Information Networks. IEEE Transactions on Multimedia (TMM) 23 (2020), 2019–2032

2020
[36]

Leheng Sheng, An Zhang, Yi Zhang, Yuxin Chen, Xiang Wang, and Tat-Seng Chua
[37]

Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Ł ukasz Kaiser, and Illia Polosukhin. 2017. Attention is All you Need. In NeurIPS, Vol. 30

2017
[38]

Tongzhou Wang and Phillip Isola. 2020. Understanding contrastive representation learning through alignment and uniformity on the hypersphere. In ICML. PMLR, 9929–9939

2020
[39]

Wenjie Wang, Fuli Feng, Xiangnan He, Xiang Wang, and Tat-Seng Chua. 2021. Deconfounded Recommendation for Alleviating Bias Amplification. In KDD. 1717–1725

2021
[40]

Xiang Wang, Xiangnan He, Meng Wang, Fuli Feng, and Tat-Seng Chua. 2019. Neural Graph Collaborative Filtering. In SIGIR. 165–174

2019
[41]

Xiang Wang, Tinglin Huang, Dingxian Wang, Yancheng Yuan, Zhenguang Liu, Xiangnan He, and Tat-Seng Chua. 2021. Learning Intents behind Interactions with Knowledge Graph for Recommendation. In WWW. 878–887

2021
[42]

Xiang Wang, Hongye Jin, An Zhang, Xiangnan He, Tong Xu, and Tat-Seng Chua
[43]

Yifan Wang, Suyao Tang, Yuntong Lei, Weiping Song, Sheng Wang, and Ming Zhang. 2020. DisenHAN: Disentangled Heterogeneous Graph Attention Nnet- work for Recommendation. In CIKM. 1605–1614

2020
[44]

Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. 2022. Chain-of-thought Prompting Elicits Reason- ing in Large Language Models. NeurIPS 35 (2022), 24824–24837

2022
[45]

Chuhan Wu, Fangzhao Wu, Yongfeng Huang, and Xing Xie. 2023. Personal- ized News Recommendation: Methods and Challenges. ACM Transactions on Information Systems (TOIS) 41, 1 (2023), 1–50

2023
[46]

Jiancan Wu, Xiang Wang, Fuli Feng, Xiangnan He, Liang Chen, Jianxun Lian, and Xing Xie. 2021. Self-supervised Graph Learning for Recommendation. In SIGIR. 726–735

2021
[47]

Yunjia Xi, Weiwen Liu, Jianghao Lin, Xiaoling Cai, Hong Zhu, Jieming Zhu, Bo Chen, Ruiming Tang, Weinan Zhang, and Yong Yu. 2024. Towards Open-world Recommendation with Knowledge Augmentation from Large Language Models. In RecSys. 12–22

2024
[48]

Yonghui Yang, Le Wu, Zihan Wang, Zhuangzhuang He, Richang Hong, and Meng Wang. 2024. Graph Bottlenecked Social Recommendation. In KDD. 3853–3862

2024
[49]

Junliang Yu, Hongzhi Yin, Xin Xia, Tong Chen, Lizhen Cui, and Quoc Viet Hung Nguyen. 2022. Are Graph Augmentations Necessary? Simple Graph Contrastive Learning for Recommendation. In SIGIR. 1294–1303

2022
[50]

Zhenrui Yue, Sara Rabhi, Gabriel de Souza Pereira Moreira, Dong Wang, and Even Oldridge. 2023. LlamaRec: Two-Stage Recommendation using Large Language Models for Ranking. CIKM (2023)

2023
[51]

Yi Zhang, Lei Sang, and Yiwen Zhang. 2024. Exploring the Individuality and Collectivity of Intents behind Interactions for Graph Collaborative Filtering. In SIGIR. 1253–1262

2024
[52]

Yi Zhang, Yiwen Zhang, Lei Sang, and Victor S Sheng. 2024. Simplify to the Limit! Embedding-less Graph Collaborative Filtering for Recommender Systems. ACM Transactions on Information Systems (TOIS) 43, 1 (2024), 1–30

2024
[2020]

In SIGIR

Disentangled Graph Collaborative Filtering. In SIGIR. 1001–1010
[2022]

arXiv preprint arXiv:2201.10005 (2022)

Text and Code Embeddings by Contrastive Pre-training. arXiv preprint arXiv:2201.10005 (2022)

2022 arXiv
[2024]

arXiv preprint arXiv:2403.03952 (2024)

Bridging language and items for retrieval and recommendation. arXiv preprint arXiv:2403.03952 (2024)

2024 arXiv
[2025]

ICLR (2025)

Language Representations Can be What Recommenders Need: Findings and Potentials. ICLR (2025)

2025

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.

[1] [1]

Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Floren- cia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 Technical Report. arXiv preprint arXiv:2303.08774 (2023)

arXiv 2023

[2] [2]

Keqin Bao, Jizhi Zhang, Yang Zhang, Wenjie Wang, Fuli Feng, and Xiangnan He. 2023. Tallrec: An Effective and Efficient Tuning Framework to Align lLarge Language Model with Recommendation. In RecSys. 1007–1014

work page 2023

[3] [3]

Xuheng Cai, Chao Huang, Lianghao Xia, and Xubin Ren. 2023. LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation. In ICLR

work page 2023

[4] [4]

Hong Chen, Yudong Chen, Xin Wang, Ruobing Xie, Rui Wang, Feng Xia, and Wenwu Zhu. 2021. Curriculum Disentangled Recommendation with Noisy Multi- feedback. NeurIPS 34 (2021), 26924–26936

work page 2021

[5] [5]

Lei Chen, Le Wu, Richang Hong, Kun Zhang, and Meng Wang. 2020. Revisiting Graph based Collaborative Filtering: A Linear Residual Graph Convolutional Network Approach. In AAAI, Vol. 34. 27–34

work page 2020

[6] [6]

Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. 2020. A Simple Framework for Contrastive Learning of Visual Representations. In ICML. PMLR, 1597–1607

work page 2020

[7] [7]

Yongjun Chen, Zhiwei Liu, Jia Li, Julian McAuley, and Caiming Xiong. 2022. Intent Contrastive Learning for Sequential Recommendation. In WWW. 2172–2182

work page 2022

[8] [8]

Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al. 2024. The Llama 3 Herd of Models. arXiv preprint arXiv:2407.21783 (2024)

arXiv 2024