Enabling Explainable Recommendation in E-commerce with LLM-powered Product Knowledge Graph

Dan Schonfeld; Duanfeng Zhang; Jianian Jin; Menghan Wang; Minnie Li; Shawn Zhou; Yuchen Guo

REVIEW 4 major objections 5 minor 2 cited by

Enabling Explainable Recommendation in E-commerce with LLM-powered Product Knowledge Graph

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A product knowledge graph distilled from an LLM raised e-commerce purchases by 7.59% in a live A/B test.

desk verdict A useful industrial recipe for distilling LLM knowledge into a product KG, but the online A/B test as designed can't pin the +7.59% transaction lift on the LLM-PKG rather than on the self-loop same-product recall or the rationale subtitles. read the letter →

arxiv 2412.01837 v1 pith:NFDDQAOO submitted 2024-11-17 cs.IR cs.LG

Menghan Wang , Yuchen Guo , Duanfeng Zhang , Jianian Jin , Minnie Li , Dan Schonfeld , Shawn Zhou This is my paper · ORCID

classification cs.IRcs.LG

keywords LLM-poweredproductknowledgegraphexplainablerecommendatione-commerceconstructionA/Btestinghallucinationpruningcandidateretrievallargelanguagemodels

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

Large language models are good at explaining why one product goes with another, but calling an LLM for every page view is too slow for e-commerce. LLM-PKG instead asks an LLM once, offline, to generate recommendation triples between seed products, turns those triples into a product knowledge graph, and then serves from that graph in real time, using each edge's predicate as the user-facing explanation. In a live A/B test on the sneaker category, the LLM-PKG treatment raised clicks by 5.19% and transactions by 7.59% relative to the existing recommender. The paper argues this shows LLM knowledge can be distilled into a practical, explainable retrieval stage that works within web-scale latency budgets. A reader should care because the approach offers a path from expensive interactive LLM recommenders to deployable systems that still carry the model's reasoning.

What carries the argument

The central object is the product knowledge graph (PKG): a directed graph whose nodes are real catalog products and user-group labels, and whose edges are (subject, predicate, object) triples where the predicate is a short rationale such as 'Same brand' or 'Classic colorway appeal'. It is built offline by prompting an LLM with a seed product and a one-shot JSON template, parsing the response into triples, then validating each edge with a scoring/reasoning LLM prompt and pruning low-quality nodes. Graph nodes are mapped to enterprise inventory by embedding product titles with a fine-tuned BERT model and running KNN vector search. At serving time, the KG is cached in a key-value store: the seed product's key returns neighboring products and their edge predicates, which are passed to a ranker and displayed as subtitles. The predicate text is thus the same artifact that explains the recommendation and connects the products.

What would settle it

Run a crossover A/B test in which the LLM-PKG candidate set is shown with control-style category-name subtitles while the control candidate set is shown with rationale subtitles; if the transaction lift follows the rationale text rather than the candidate source, the paper's causal story about the knowledge graph is wrong. Removing self-loop edges from the treatment group and observing whether the 7.59% uplift persists is a second direct check.

Watch

Extended reading notes

Core claim

LLM-PKG is a framework that distills a large language model's product knowledge into an offline knowledge graph and uses that graph as the candidate-recall stage of an e-commerce recommender. In the construction phase, prompted LLM outputs are parsed into (subject, predicate, object) triples that form a product-level graph; the graph is then pruned and its edge rationales rewritten by LLM-based evaluation, and its nodes are mapped to real catalog items through BERT-embedding KNN search. Online, cached triples let the item page fetch nearby products and display the edge predicate as a short rationale. The paper's central claim is that serving these LLM-PKG candidates with their rationale subtitles significantly outperforms the existing control — collaborative filtering plus title similarity with category-name subtitles — in a controlled A/B test on the sneaker category: clicks rose 5.19%, transactions (1-day maturity) rose 7.59%, CTR rose 5.91%, and GMB rose 8.56%. The authors interpret this as evidence that both the LLM's product understanding and the displayed explanations drive engagement and purchases.

Load-bearing premise

The paper's conclusion assumes that the +7.59% transaction and +5.19% click lifts are caused by the LLM-PKG knowledge graph and its rationale subtitles, not by other differences between the treatment and control groups — most notably the self-loop 'same product' recall and the different subtitle text.

Editorial extensions

If this is right

LLM knowledge can be reused across many sessions without paying real-time inference cost, making explainable LLM-based recommendations feasible at web scale.
Recommendation explanations require no separate generation step: the edge predicate used for retrieval is directly shown to the user.
The same graph structure supports both item-based recommendations (product-to-product edges) and user-based recommendations (user-group edges), with user-based serving left as future work.
The validation and pruning loop, with LLM scores and rationale rewriting, provides a practical checklist for keeping LLM hallucination in check during KG construction.

Reading between the lines

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

The A/B test varies two things at once — candidate recall source and subtitle text — so the isolated contribution of the rationale text is not identified; an ablation that crosses these factors would tell whether the lift comes from better candidates or from the explanation on screen.
The treatment's permitted self-loop edges (same product from a different seller) may inflate conversion metrics, since these recommendations are essentially direct substitutes; re-running without self-loops would test whether the uplift reflects genuinely novel discovery.
The vector-search mapping from LLM output to catalog items could create coverage gaps for products with no close embedding neighbor; measuring what fraction of seed products have no valid KG edges would expose this.
If the mechanism is truly the LLM's world knowledge, the approach should transfer to categories with richer relational structure; a comparable test in apparel or electronics would be a natural next experiment.

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

Colleague,

The paper is a good example of industrial LLM+KG engineering, but the headline claim is not established by the experiment as reported. Here is the take: the offline pipeline (LLM-generated RDF triples, validation and pruning with the same LLM, vector-search mapping to real products, key-value caching for serving) is a sensible, cost-effective way to get LLM knowledge into a recommender without real-time LLM calls. That part is credible and worth reading.

The new result is the specific end-to-end system plus a live A/B test on the sneaker category: +5.19% clicks and +7.59% 1-day transactions, with p-values. That is a real data point, and the writeup is honest about the graph statistics (355k nodes, 27.86% relation imprecise rate before pruning). Credit where due: they report enterprise-scale numbers, they show the prompts, and they give a concrete serving design.

Now the soft spots. The A/B test is confounded. The treatment differs from control in at least three ways at once: recall source (LLM-PKG vs collaborative filtering/title similarity), subtitle text (rationale vs category name), and self-loop edges (same product from different sellers). The self-loop is not a minor detail. On a View Item Page after 'Add to Watchlist', showing the exact same product from another seller with a 'Same Product' subtitle can plausibly increase clicks and transactions by itself, with zero LLM knowledge or explanation involved. The treatment also falls back to control for seeds absent from the PKG, so the reported lift is a blended effect over eligible and fallback traffic. There is no ablation, no confidence intervals, no per-edge or per-subtitle breakdown. As a result, the paper's abstract sentence about 'driving user engagements and transactions significantly' is safe, but the attribution to LLM-PKG and its rationales is not supported.

I disagree with the reader on one minor point: the self-referential LLM validation (LLM scores its own edges) is a real concern for KG quality, but it is not a fatal flaw because the final benchmark is real user behavior. The circularity would matter more if the only evaluation were LLM preferences.

Who is this for? Practitioners in e-commerce or industrial recommender systems who want a template for offline LLM distillation. It is a methods paper, not a scientific breakthrough. It deserves a serious referee because the system is concrete, the experiment is real, and the confounds are fixable with a better design. I would ask for a disentangled experiment before trusting the causal claim.

Recommendation: send to peer review, with a request for an ablation that separates self-loop, rationale text, and recall source, plus confidence intervals.

Referee Report

4 major / 5 minor

Summary. The paper proposes LLM-PKG, an offline pipeline that uses a large language model to generate a product knowledge graph (PKG) from curated prompts, maps graph nodes to real enterprise inventory via KNN search with fine-tuned BERT embeddings, validates and prunes the graph with LLM-based scoring, and serves explainable item-based and user-based recommendations from the resulting graph. The authors evaluate LLM-PKG in an online A/B test on the sneaker category of an e-commerce site, comparing the treatment (LLM-PKG recall plus LLM-generated rationale subtitles, with self-loop edges allowed) against a control (collaborative filtering and title-similarity recall with fixed category-name subtitles). They report statistically significant relative lifts in clicks (+5.19%), transactions at 1-day maturity (+7.59%), and other metrics, concluding that LLM-PKG drives user engagement and transactions significantly.

Significance. If the causal attribution were established, the paper would describe a practical, latency-friendly way to inject LLM knowledge into a production recommendation system and to provide explanations at scale. The offline construction and product-mapping pipeline is clearly presented, and the deployment on live traffic with several engagement and revenue metrics is an unusual and welcome strength. However, the experimental design conflates several factors—candidate recall source, subtitle text, and permission of self-loop edges—so the reported lifts cannot be uniquely attributed to the LLM-powered knowledge graph or to its explainability. The paper's central claim therefore overstates what the evidence supports; the results demonstrate only that the combined treatment package outperformed the control. The significance, as currently worded, is thus proportional to a stronger experimental design than the one presented.

major comments (4)

[Sec. 4.1 Experimental Design] The treatment and control differ in at least three simultaneous aspects: (1) candidate recall uses LLM-PKG versus collaborative filtering/title similarity, (2) displayed subtitles are LLM-generated rationales versus fixed category names, and (3) the treatment explicitly permits self-loop edges that recall the same product from different sellers, whereas the control has no equivalent. Because no ablation or per-edge analysis is reported, the +7.59% transaction lift and other improvements in Table 3 cannot be attributed specifically to the LLM-generated knowledge or to the explanation subtitles. In particular, the self-loop edges may mechanically increase clicks and transactions by surfacing competing offers for the same product after a user clicks 'Add to Watchlist', which is a plausible behavior without any contribution from LLM knowledge. An ablation that isolates the self-loop edges, or a per-edge-type breakdown of engagement, is needed to support the central claim.
[Sec. 4.1 Experimental Design (fallback)] The treatment falls back to the control whenever the seed item has no corresponding node in the LLM-PKG, so the reported treatment effect is a blended average over traffic that actually received LLM-PKG candidates and traffic that received the control. The paper does not report the fraction of treatment impressions that received LLM-PKG recommendations, nor the effect conditional on receiving them. If, for example, the LLM-PKG coverage of the sneaker seed set is low, the true effect on exposed users could be substantially larger than the reported blended lift, or conversely the blended result could be dominated by a confounded subset. Reporting coverage statistics and a per-segment analysis of treated versus fallback traffic is necessary to interpret the magnitude of the lift.
[Sec. 4.2 Results, Table 3] All metrics are reported as relative differences with p-values, without confidence intervals, sample sizes, or absolute baseline rates. Relative lifts of 5–8% on unbounded baselines can be practically negligible or substantial, and with the confounded design the precision of the estimate is central to the claim of 'significant' improvement. I request at least 95% confidence intervals and the traffic volumes or event counts underlying each metric. This is not merely a presentation issue: it determines whether the observed lift is robust enough to ground the paper's conclusion.
[Sec. 3.3 Validate and Refine LLM-PKG] The graph is validated and pruned using the same family of LLMs that generated the graph: the LLM assigns acceptability scores to its own recommended products and judges the accuracy of its own rationales. While the online A/B test is an external benchmark and therefore not entirely circular, the offline quality metrics in Table 2 (e.g., Avg Edge Score 8.57, Relation Imprecise Rate 27.86% before pruning) are self-assessments and may not reflect human-perceived quality. The paper also does not explain how the pruning threshold of 6 was chosen or how sensitive the final graph is to this free parameter. I ask the authors to acknowledge this limitation explicitly and, if possible, report a small human evaluation of a random sample of graph edges to ground the qualitative claims about quality.

minor comments (5)

[Sec. 3.5 Online Serving] The text says 'leave the user-centric recommendations for furture work'—'furture' should be 'future'.
[Sec. 4.1 Experimental Design] The phrase 'repsectively' is a typo for 'respectively'.
[Table 1 / Table 4] There are small grammatical errors in the prompt examples: 'A user of a eBay' should be 'of eBay' or 'of an eBay user', and 'F or' appears in the task description where 'For' is intended.
[Sec. 3.2 Construction of LLM-PKG] The prompt template in Table 1 includes a JSON format example where the seed product node is written as {"product title", seed product, ...}, but the output example in Table 4 uses a dict-like structure; please make the format indicators consistent across both tables.
[Table 2] The table columns mix 'Initial Graph' and 'Enterprise Product Graph' statistics; the 'Avg Edge Score' and 'Relation Imprecise Rate' are labeled as 'Before pruning', but the table is already showing results after some refinement (per the text). Please clarify which stage each column describes and define 'Relation Imprecise Rate' precisely.

Circularity Check

1 steps flagged · score 1.0 of 10

No significant circularity: the headline A/B lift is an external, real-user benchmark; only the internal LLM self-evaluation/pruning loop is self-referential, and it is not load-bearing for the reported transaction lift.

self definitional [Section 3.3 (Validate and Refine LLM-PKG)]
"As LLM has excellent role-play capabilities we ask LLM to evaluate the quality of nodes and edges in the KG with prompts. ... We prompt LLM to assign a score ranging from 1 to 10 to each seed product and recommended product pair based on their acceptability and relevance. ... We then use the score to prune nodes."

The same LLM is both the generator of the recommendation edges and the judge that scores those edges; the pruning threshold is applied to scores produced by the same model distribution that produced the edges. Consequently, the 'quality' and 'reliability' of the final KG are defined by the LLM's self-assessment rather than by independent ground truth, so the purported hallucination-mitigation step does not provide independent validation. This loop is, however, separate from the online A/B test, which measures real user behavior, so it does not by itself force the reported +7.59% transaction lift.

full rationale

The headline claim—that the LLM-PKG treatment produced +5.19% clicks and +7.59% 1-day-maturity transactions (Table 3)—is supported by an online A/B test against the existing recommender, an external outcome measure that does not depend on the LLM's own scores; therefore the central derivation is not circular. The only genuine self-referential loop is in Sec. 3.3, where the same LLM generates and then scores/prunes the graph edges, so the 'validation' is a self-assessment rather than an independent check; this weakens the hallucination-mitigation claim but is not load-bearing for the A/B lift. The multiple simultaneous differences between treatment and control (LLM-PKG recall, rationale subtitles, and permitted self-loop edges) are experimental confounds and attribution limitations, not circularity.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The system rests on a chain of unverified assumptions: LLM world knowledge, LLM self-evaluation, embedding-space product mapping, and clean A/B attribution. Only a single online experiment connects the pipeline to user behavior, and it conflates multiple changes.

free parameters (2)

pruning score threshold = 6
Nodes and edges with LLM acceptability score below 6 are removed; the threshold is chosen by hand with no ablation study.
number of recommendations per seed (k) = 5 in the example prompt
The prompt requests k recommendations; the value is fixed without sensitivity analysis.

assumptions (4)

domain assumption LLM possesses reliable world knowledge about product use cases and purchase intentions.
Stated in the Introduction: 'LLM is assumed to own world knowledge.' This underpins the entire KG construction.
domain assumption LLM role-play capabilities are sufficient to judge the quality of its own generated recommendations.
Section 3.3 uses the LLM as evaluator for scoring and pruning; no independent human validation is provided.
ad hoc to paper KNN vector search with fine-tuned BERT embeddings faithfully maps LLM-generated product titles to real inventory products.
Section 3.4 assumes semantic similarity in embedding space corresponds to correct product identity, with no accuracy benchmark for the mapping.
domain assumption A/B test users are independent and the traffic split is unbiased.
Section 4.1 states 'traffic allocation between the Treatment and Control groups was evenly distributed' but gives no details on user overlap or randomization.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enabling Explainable Recommendation in E-commerce with LLM-powered Product Knowledge Graph." pith.science (2026). https://pith.science/paper/NFDDQAOO

@misc{pith2026241201837,
  author       = {Pith},
  title        = {Pith review of: Enabling Explainable Recommendation in E-commerce with LLM-powered Product Knowledge Graph},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NFDDQAOO}},
  note         = {Machine review of arXiv:2412.01837}
}

read the original abstract

How to leverage large language model's superior capability in e-commerce recommendation has been a hot topic. In this paper, we propose LLM-PKG, an efficient approach that distills the knowledge of LLMs into product knowledge graph (PKG) and then applies PKG to provide explainable recommendations. Specifically, we first build PKG by feeding curated prompts to LLM, and then map LLM response to real enterprise products. To mitigate the risks associated with LLM hallucination, we employ rigorous evaluation and pruning methods to ensure the reliability and availability of the KG. Through an A/B test conducted on an e-commerce website, we demonstrate the effectiveness of LLM-PKG in driving user engagements and transactions significantly.

Figures

Figures reproduced from arXiv: 2412.01837 by the authors.

**Figure 1.** Overall Framework • Scoring Recommended Products: We adopt a scoring mechanism to assess the relative quality of each recommended product. We prompt LLM to assign a score ranging from 1 to 10 to each seed product and recommended product pair based on their acceptability and relevance. This scoring mechanism serves as a quantitative measure of the suitability of recommended items within the context of user prefere… view at source ↗

**Figure 2.** Effects of One-shot learning. Metrics Initial Graph Enterprise Product Graph Nodes Edges Avg Edge Score (Before pruning) Relation Imprecise Rate (Before pruning) Nodes Edges Value 355,383 788,974 8.57 27.86% 325,721 718,365 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Sneaker LLM-PKG Visualization (Sampled) Report Metrics Difference P-Value Clicks +5.19% 0.0 Transactions (1-Day Maturity) +7.59% 0.0007 Transactions (7-Day Maturity) +7.55% 0.0 Click Through Rate +5.91% 0.0 Conversion Rate +7.20% 0.0038 GMB (1-Day Maturity) +8.56% 0.0459 Ads Revenue (1-Day Maturity) +10.84% 0.0184 [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: An example of LLM-PKG based recommendation in ex [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

Graph Foundation Models for Recommendation: A Comprehensive Survey
cs.IR 2025-02 conditional novelty 4.0 of 10

A comprehensive survey that categorizes graph foundation model approaches to recommendation into graph-augmented LLM, LLM-augmented graph, and LLM-graph harmonization.
Large Language Model Enhanced Recommender Systems: A Survey
cs.IR 2024-12 unverdicted novelty 4.0 of 10

A survey organizing LLM-enhanced recommender systems into knowledge, interaction, and model enhancement, and tracing a shift from explicit text to implicit embeddings and fine-tuned open-source LLMs.

Reference graph

Works this paper leans on

9 extracted references · 1 canonical work pages · cited by 2 Pith papers

[1]

Knowledge graphs as context sources for llm-based explanations of learning recommendations

Hasan Abu-Rasheed, Christian Weber, and Madjid Fathi. Knowledge graphs as context sources for llm-based explanations of learning recommendations. arXiv preprint arXiv:2403.03008 , 2024

arXiv 2024
[2]

Chat-rec: Towards interactive and explainable llms-augmented recommender system

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

arXiv 2023
[3]

Large language models are zero-shot rankers for recommender systems

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

work page 2024
[4]

Is chatgpt a good recommender? a preliminary study

Junling Liu, Chao Liu, Renjie Lv, Kang Zhou, and Yan Zhang. Is chatgpt a good recommender? a preliminary study. arXiv preprint arXiv:2304.10149 , 2023

arXiv 2023
[5]

Kellmrec: Knowledge-enhanced large language models for recommendation

Weiqing Luo, Chonggang Song, Lingling Yi, and Gong Cheng. Kellmrec: Knowledge-enhanced large language models for recommendation. arXiv preprint arXiv:2403.06642 , 2024

arXiv 2024
[6]

Lkpnr: Llm and kg for personalized news recommendation framework

Xie Runfeng, Cui Xiangyang, Yan Zhou, Wang Xin, Xuan Zhanwei, Zhang Kai, et al. Lkpnr: Llm and kg for personalized news recommendation framework. arXiv preprint arXiv:2308.12028 , 2023

arXiv 2023
[7]

Knowledge graph large language model (kg-llm) for link prediction

Dong Shu, Tianle Chen, Mingyu Jin, Yiting Zhang, Mengnan Du, and Yongfeng Zhang. Knowledge graph large language model (kg-llm) for link prediction. arXiv preprint arXiv:2403.07311 , 2024

arXiv 2024
[8]

Recmind: Large language model powered agent for recommendation

Yancheng Wang, Ziyan Jiang, Zheng Chen, Fan Yang, Yingxue Zhou, Eunah Cho, Xing Fan, Xiaojiang Huang, Yanbin Lu, and Yingzhen Yang. Recmind: Large language model powered agent for recommendation. arXiv preprint arXiv:2308.14296 , 2023

arXiv 2023

Show all 9 references

[9]

write newline

" write newline "" before.all 'output.state := FUNCTION fin.entry add.period write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence '...

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Knowledge graphs as context sources for llm-based explanations of learning recommendations

Hasan Abu-Rasheed, Christian Weber, and Madjid Fathi. Knowledge graphs as context sources for llm-based explanations of learning recommendations. arXiv preprint arXiv:2403.03008 , 2024

arXiv 2024

[2] [2]

Chat-rec: Towards interactive and explainable llms-augmented recommender system

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

arXiv 2023

[3] [3]

Large language models are zero-shot rankers for recommender systems

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

work page 2024

[4] [4]

Is chatgpt a good recommender? a preliminary study

Junling Liu, Chao Liu, Renjie Lv, Kang Zhou, and Yan Zhang. Is chatgpt a good recommender? a preliminary study. arXiv preprint arXiv:2304.10149 , 2023

arXiv 2023

[5] [5]

Kellmrec: Knowledge-enhanced large language models for recommendation

Weiqing Luo, Chonggang Song, Lingling Yi, and Gong Cheng. Kellmrec: Knowledge-enhanced large language models for recommendation. arXiv preprint arXiv:2403.06642 , 2024

arXiv 2024

[6] [6]

Lkpnr: Llm and kg for personalized news recommendation framework

Xie Runfeng, Cui Xiangyang, Yan Zhou, Wang Xin, Xuan Zhanwei, Zhang Kai, et al. Lkpnr: Llm and kg for personalized news recommendation framework. arXiv preprint arXiv:2308.12028 , 2023

arXiv 2023

[7] [7]

Knowledge graph large language model (kg-llm) for link prediction

Dong Shu, Tianle Chen, Mingyu Jin, Yiting Zhang, Mengnan Du, and Yongfeng Zhang. Knowledge graph large language model (kg-llm) for link prediction. arXiv preprint arXiv:2403.07311 , 2024

arXiv 2024

[8] [8]

Recmind: Large language model powered agent for recommendation

Yancheng Wang, Ziyan Jiang, Zheng Chen, Fan Yang, Yingxue Zhou, Eunah Cho, Xing Fan, Xiaojiang Huang, Yanbin Lu, and Yingzhen Yang. Recmind: Large language model powered agent for recommendation. arXiv preprint arXiv:2308.14296 , 2023

arXiv 2023