Pith. sign in

REVIEW 3 major objections 6 minor 30 references

GenRec: An LLM-Backed Recommendation Ranker at Netflix

T0 review · 3 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read An LLM-backed ranker that reads user histories as text beats a mature production recommender while using roughly 1/40th of the Phase-2 labeled data.

desk verdict Genuine industrial-scale evidence that an LLM-backed ranker can beat a mature production baseline, but the paper must disclose whether its long-term metric overlaps with its training rewards. read the letter →

arxiv 2608.10257 v1 pith:LPZMBQAX submitted 2026-08-10 cs.IR

Ying Li , Shradha Sehgal , Arjun Rao , Rein Houthooft , Yaochen Zhu , Ashish Rastogi This is my paper · ORCID
classification cs.IR
keywords largelanguagemodelsrecommendationrankingpost-trainingcontextengineeringreward-weightedlosscatalog-awarescoringprefill-onlyinferenceA/Btesting
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

GenRec is Netflix's attempt to replace a mature, heavily engineered recommendation ranker with a large language model that reads user histories and context as natural language. The paper's central claim is that this LLM-backed ranker, post-trained in a second phase on roughly $40\times$ fewer labeled examples and far fewer input signals than the production model, still wins: about +1.6% relative offline MRR and statistically significant gains in a four-week A/B test on 10% of traffic, specifically +0.115% on a short-term engagement metric and +0.006% on a long-term core metric. The authors argue this marks a shift from feature engineering to context engineering, and from bespoke ranking architectures to shared foundation backbones. A sympathetic reader would care because it suggests that LLM-native ranking can be both data-efficient and cost-viable at industrial scale, not just a research curiosity.

What carries the argument

The load-bearing object is the catalog-aware ranking head mounted on a decoder-only LLM backbone. A verbalizer $V$ turns history $H$, item metadata $\{M_i\}_{i \in C}$, and context $\tau$ into a text sequence $x$; the LLM produces a pooled representation $h$; a scoring head $\phi$ combines $h$ with learned item embeddings $e_i$ to score every catalog item, with a softmax over the catalog giving the ranking. This design avoids autoregressive decoding over items, so large candidate sets can be ranked in one forward pass. Around it sit two supporting mechanisms: a reward-weighted ranking loss that scales each training example by reward-model scores to steer toward long-term satisfaction and business constraints, and prefill-only inference on the serving stack so cost scales with one context pass rather than token-by-token generation. The paper also treats context engineering as a first-class lever, compressing the prompt to roughly a third of its original token budget with negligible offline degradation.

What would settle it

Compute the correlation between the reward-model scores used to weight Phase-2 training examples and the long-term core metric reported in the A/B test; if the correlation is high, or if retraining GenRec without reward weights eliminates the +0.006% long-term lift, the central claim that GenRec improves long-term member satisfaction is undermined.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that a generative, decoder-only LLM augmented with a catalog-aware scoring head can outperform a production discriminative ranker tuned over many years. GenRec verbalizes member interaction histories, item metadata, and context into a prompt; the LLM encodes the prompt once, and a scoring head ranks the full catalog in a single prefill-only forward pass. With roughly $40\times$ fewer Phase-2 labeled training examples and substantially fewer input signals, GenRec reports about +1.6% relative MRR offline, and online A/B results of +0.115% on a short-term homepage engagement metric and +0.006% on a long-term core metric, both statistically significant. The paper also reports that Phase-1 foundation training contributes roughly 10–20% offline MRR over an off-the-shelf LLM, Phase-2 post-training adds 35–50% (rising to about 80% as the Phase-1 backbone ages), and careful verbalization compaction cuts context length to about a third with negligible quality loss.

Load-bearing premise

The result depends on the long-term online metric being a genuine measure of member satisfaction that is independent of the reward-model scores used to weight GenRec's training examples; if the two are the same or highly correlated, the reported long-term gain is an artifact of training toward the metric rather than evidence of improved satisfaction.

Editorial extensions

If this is right

  • If the central claim holds, a production recommender can drop most hand-engineered features and bespoke multi-task architectures in favor of a shared LLM backbone that reads verbalized context.
  • GenRec's Phase-2 data efficiency (about $40\times$ fewer labeled examples) means frequent ranker refreshes become much cheaper, since Phase-2 is the high-cadence stage.
  • Context compaction from roughly 5,000 to 1,700 tokens cuts serving cost to about a third with negligible quality loss, making LLM ranking affordable at high traffic.
  • The scaling curves reported in the paper—monotonic gains with more Phase-2 data and with larger backbones—imply that quality–cost trade-offs can be planned with data and model size, not only with architecture tricks.
  • A catalog-aware scoring head plus softmax over the catalog means out-of-catalog recommendations are structurally impossible, which addresses a known LLM failure mode.

Reading between the lines

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

  • Editorial inference: the same architecture should transfer to other recommendation surfaces and content types at Netflix, because the prompt, not the model, carries surface-specific context; the main risk is that catalog-aware scoring over a very large catalog may need sampled softmax, which could change the quality–cost curve.
  • Editorial inference: if the reward-model scores used to weight Phase-2 examples align closely with the long-term core metric in the A/B test, the reported long-term lift may partly be an artifact of optimizing the training objective toward the metric; the paper does not disclose the relationship, so this needs a direct test.
  • Editorial inference: a testable extension is to ablate reward weighting in a second A/B test, comparing a version with zero reward weights against the full GenRec model; if the long-term lift disappears, reward alignment is the active ingredient, not the LLM backbone per se.
  • Editorial inference: because the language modeling objective is kept during training but not used at inference, GenRec leaves the door open to future natural-language outputs such as explanations or steerable prompts; that capability is not demonstrated here.
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

3 major / 6 minor

Summary. The paper describes GenRec, a large-language-model-backed recommendation ranker developed at Netflix, and evaluates it against a mature production ranker. The system verbalizes user histories, item metadata, and context into prompts for an in-house foundation LLM, augments the decoder-only backbone with a catalog-aware scoring head for single-pass ranking, and post-trains with a multi-objective loss that combines ranking, language modeling, and reward-weighted examples. The central empirical claim is that a GenRec model trained with roughly 40x fewer Phase-2 labeled examples and far fewer input signals achieves a +1.6% relative offline MRR gain and statistically significant online gains on short-term (+0.115%) and long-term (+0.006%) metrics in a four-week, 10%-traffic A/B test. The paper also reports data and model scaling trends, a Phase 1 versus Phase 2 decomposition, context-compaction results, and serving-cost optimizations based on prefill-only inference.

Significance. If the result holds, this is an important industrial data point for LLM-native recommendation: it compares against a long-running production baseline, uses a large-scale online A/B test, and provides concrete evidence that an LLM-backed ranker can be competitive while simplifying feature engineering. The Phase 1/Phase 2 decomposition and context-length ablations are practically useful, and the paper is honest about its scope (batch-compute surfaces, low-data/low-signal configuration, and the fact that RL-style alignment is left for future work). The central unresolved issue is the relationship between the reward models used in training and the long-term evaluation metric; without that disclosure, the long-term component of the headline claim is not independently verifiable.

major comments (3)
  1. [§4.6, §5.1 (Figure 3)] The long-term component of the central claim is not currently falsifiable. Section 4.6 says that reward models provide 'long-term satisfaction proxies' learned from historical data and that GenRec's training examples are weighted by these reward signals, while Section 5.1 and Figure 3 report an improvement on an undefined 'long-term core metric.' The paper never states whether this core metric is identical to, derived from, or independent of the reward-model scores. If the two coincide or are highly correlated, the +0.006% long-term lift is the expected consequence of optimizing the reward-weighted loss rather than independent evidence of improved long-term member satisfaction. Because the abstract and Section 3 motivate GenRec by alignment with long-term member utility, the authors should define the long-term core metric, state its relationship to the reward models, and ideally report a long-term measure that was not used in training.
  2. [§5.1, §5.2, §5.3 (Figures 4-5, Table 1)] Offline results are reported only as point estimates. The +1.6% relative MRR lift in Section 5.1, the monotonic scaling curves in Figure 4, the Phase 1/Phase 2 contributions in Table 1, and the context-length elbow in Figure 5 are all presented without confidence intervals, standard errors, or the number of users and requests in the offline evaluation. The +1.6% figure is the primary offline justification, but the reader cannot tell whether it is statistically distinguishable from zero or whether the apparent monotonicity in Figure 4 is within noise. Please provide confidence intervals or standard errors and the evaluation-set sizes for each offline experiment.
  3. [§4.4, §4.6] Several load-bearing design choices are selected through offline experiments without any sensitivity analysis: the multi-objective weights (alpha, beta, gamma) in Section 4.4, the label-denoising thresholds per content type, the context retention and compression thresholds in Section 4.3, and the reward scalarization parameters in Section 4.6. Because these choices directly affect the reward-weighted loss and the reported offline gains, it is unclear how robust the central offline lift is to reasonable variations in these hyperparameters. A sensitivity study, or at minimum the chosen values and the range over which the reported gains remain positive, would strengthen the empirical claim.
minor comments (6)
  1. [Figure 3] The y-axis labels in Figure 3 ("0 (significant)") are confusing; please show point estimates with confidence intervals and clarify what the zero baseline denotes.
  2. [§4.4] The constraint alpha + beta + gamma = 1 with nonnegative weights leaves only two degrees of freedom; please state whether this is a normalization convention or a substantive modeling choice.
  3. [§4.5] Please specify how the pooled representation h is obtained (which layer and which pooling position) and define the dimension d explicitly.
  4. [§5.3] The statement that the Phase-2 gain 'rises to roughly 80% after two weeks' needs a definition of the evaluation protocol: is this offline MRR measured at a fixed cutoff, and how is the increasing staleness of the Phase-1 model operationalized?
  5. [§4.7] The claim that serving cost is approximately proportional to context length should be justified for the prefill-only configuration, since prefix caching, batching, and the catalog-aware scoring head may change the scaling behavior.
  6. [§7] There is a typo in the conclusion: 'Phase2-2 labels' should be 'Phase-2 labels'; also fix the spacing in 'to athird' in Section 5.4.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: GenRec's claims rest on offline and online empirical comparisons against an external production baseline, with no exhibited definitional or self-citation reduction.

full rationale

The paper's central claim is an empirical comparison: a Phase-2 post-trained LLM ranker is evaluated offline (MRR) and online (A/B on 10% traffic for 4 weeks) against the production ranker. MRR is an independent held-out ranking metric, and the online metrics are external business metrics; neither is defined in terms of GenRec's training objective. The training pipeline (Sections 4.4-4.6) uses labels and reward-weighted losses, but the reported gains (Section 5.1, Figure 3) are not derived from those quantities by construction; they come from a live experiment against a mature baseline. The Phase 1/Phase 2 ablation (Table 1), data-scaling curve (Figure 4), and context-length sweep (Figure 5) are also empirical measurements, not tautologies. One legitimate ambiguity is the relation between the reward models in Section 4.6 and the 'long-term core metric' in Section 5.1; if they coincided, the long-term lift would partly reflect training toward the metric. However, the paper never defines either object, so the specific reduction required by the circularity standard is not exhibited in the text. Without evidence that the evaluation metric equals the reward-model target, this remains a disclosure or correctness concern, not demonstrated circularity. No load-bearing self-citation chain is present; reference [22] supplies reward models but is not used to justify the central claim.

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

The central claim is empirical, not derived, so the ledger records design choices and assumptions that affect the outcome. Loss weights, label thresholds, context-selection thresholds, and reward weighting are tuned on data (or hand-chosen), so they are free parameters. The main assumptions are the strength of the production baseline, the validity of MRR as a proxy, the accuracy of reward models, and the sufficiency of prefill-only scoring.

free parameters (4)
  • Multi-objective loss weights (alpha, beta, gamma) = not specified (tuned via offline experiments)
    Equation in Section 4.4 balances ranking, language, and miscellaneous losses; these tuning choices affect the final model and are selected using offline metrics.
  • Label denoising thresholds per content type = not specified; vary by content type
    Section 4.4: positive labels defined by high-duration plays and explicit feedback with thresholds that vary by content type; thresholds are set by hand.
  • Event retention and compression thresholds in context engineering = elbow point in Figure 5 selected from data; concrete thresholds not given
    Section 4.3 and 5.4: high-signal engagements are retained, low-signal omitted, repetitive behavior compressed; the history-length elbow is chosen from offline sweeps.
  • Reward-weighting scalarization parameters = not specified
    Section 4.6: multiple reward model outputs are combined into a scalar per-example weight; the combination rule and scaling are not described.
assumptions (4)
  • domain assumption The production baseline is a mature, competitive ranker, so beating it is meaningful.
    Section 5.1: 'a mature discriminative model tuned over many years'; if the baseline were artificially weak, the headline claim would not hold.
  • domain assumption Offline MRR is a valid proxy for ranking quality and predicts the online metrics used in the A/B test.
    Section 5.1 and 5.2 report MRR as the main offline metric; the paper does not establish the correlation between MRR and the online metrics.
  • domain assumption The reward models from the existing framework [22] provide unbiased estimates of long-term member satisfaction.
    Section 4.6: reward-weighted ranking loss relies on these reward models; if they are biased, GenRec's training target is biased.
  • domain assumption A single pooled representation from the LLM suffices to score all catalog items; no sequential decoding is needed.
    Section 4.5 defines the scoring head using the hidden state at a pooling position; the paper's serving cost savings depend on this being sufficient.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GenRec: An LLM-Backed Recommendation Ranker at Netflix." pith.science (2026). https://pith.science/paper/LPZMBQAX

@misc{pith2026260810257,
  author       = {Pith},
  title        = {Pith review of: GenRec: An LLM-Backed Recommendation Ranker at Netflix},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LPZMBQAX}},
  note         = {Machine review of arXiv:2608.10257}
}
read the original abstract

Large language models (LLMs) are reshaping recommender systems by enabling richer modeling of users, content, and context directly in natural language. At Netflix, we are exploring this direction through GenRec, an LLM-backed recommendation ranker built on top of an in-house foundational LLM. GenRec follows a two-phase framework: Phase 1 adapts an open-source LLM to Netflix data, developing deep understanding of the catalog and member behavior while balancing capabilities such as content understanding and instruction following. Phase 2 post-trains this foundation model with recommendation-ranking specific data, labels, and reward signals, aiming to align the ranker with business requirements and long-term member satisfaction. This paper focuses on Phase 2 and the transition from a traditional discriminative ranker with thousands of engineered features to an LLM-backed ranker driven by verbalized user histories and context. We describe our design for input verbalization and context engineering, post-training data construction, reward integration, model architecture, and a cost-constrained serving design based on a prefill-only inference approach. We report results from a large-scale A/B test comparing GenRec against the current production ranker model, where we show that a GenRec model trained with substantially fewer Phase-2 labeled training examples and input signals can achieve statistically significant gains in offline and online metrics. We discuss how LLM-backed recommenders could shift the recommendation paradigm: from feature engineering to context engineering, and from bespoke architectures to shared foundation backbones. We also outline practical lessons for serving such systems under real-world resource constraints.

Figures

Figures reproduced from arXiv: 2608.10257 by the authors.

Figure 1
Figure 1. GenRec inference pipeline. Raw logs of user history, item metadata, and context are transformed via context [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Two-phase training framework. Phase 1 trains a [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Online metrics of GenRec vs. production model. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Normalized offline ranking metric (MRR) vs. num [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 4 canonical work pages

  1. [1]

    Prabhat Agarwal, Anirudhan Badrinath, Laksh Bhasin, Jaewon Yang, Edoardo Botta, Jiajing Xu, and Charles Rosenberg. 2026. PinRec: Outcome-Conditioned, Multi-Token Generative Retrieval for Industry-Scale Recommendation Systems. arXiv:2504.10507 [cs.IR] https://arxiv.org/abs/2504.10507

    arXiv 2026

  2. [2]

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

  3. [3]

    Heng-Tze Cheng, Levent Koc, Jeremiah Harmsen, Tal Shaked, Tushar Chandra, Hrishi Aradhye, Glen Anderson, Greg Corrado, Wei Chai, Mustafa Ispir, Rohan Anil, Zakaria Haque, Lichan Hong, Vihan Jain, Xiaobing Liu, and Hemal Shah

  4. [4]

    Paul Covington, Jay Adams, and Emre Sargin. 2016. Deep Neural Networks for YouTube Recommendations. InProceedings of the 10th ACM Conference on Recommender Systems(Boston, Massachusetts, USA)(RecSys ’16). Association for Computing Machinery, New York, NY, USA, 191–198. doi:10.1145/2959100. 2959190

    doi:10.1145/2959100 2016

  5. [5]

    Thomas, Alexandra Ranieri, Matthew N

    Edoardo D’Amico, Marco De Nadai, Praveen Chandar, Divita Vohra, Shawn Lin, Max Lefarov, Paul Gigioli, Gustavo Penha, Ilya Kopysitsky, Ivo Joel Senese, Darren Mei, Francesco Fabbri, Oguz Semerci, Yu Zhao, Vincent Tang, Brian St. Thomas, Alexandra Ranieri, Matthew N. K. Smith, Aaron Bernkopf, Bryan Leung, Ghazal Fazelnia, Mark VanMiddlesworth, Timothy Chris...

    arXiv 2026

  6. [6]

    Jiaxin Deng, Shiyao Wang, Kuo Cai, Lejian Ren, Qigen Hu, Weifeng Ding, Qiang Luo, and Guorui Zhou. 2025. OneRec: Unifying Retrieve and Rank with Genera- tive Recommender and Iterative Preference Alignment. arXiv:2502.18965 [cs.IR] https://arxiv.org/abs/2502.18965

    arXiv 2025

  7. [7]

    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). InProceedings of the 16th ACM Conference on Recommender Systems(Seattle, WA, USA)(RecSys ’22). Association for Computing Machinery, New York, NY, USA, 299–315. doi:10.11...

    arXiv 2022

  8. [8]

    Ruining He, Lukasz Heldt, Lichan Hong, Raghunandan Keshavan, Shifan Mao, Nikhil Mehta, Zhengyang Su, Alicia Tsai, Yueqi Wang, Shao-Chuan Wang, Xinyang Yi, Lexi Baugher, Baykal Cakici, Ed Chi, Cristos Goodrow, Ningren Han, He Ma, Romer Rosales, Abby Van Soest, Devansh Tandon, Su-Lin Wu, Weilong Yang, and Yilin Zheng. 2026. PLUM: Adapting Pre-trained Langua...

    arXiv 2026

Show all 30 references
  1. [9]

    Yupeng Hou, An Zhang, Leheng Sheng, Zhengyi Yang, Xiang Wang, Tat-Seng Chua, and Julian McAuley. 2025. Generative Recommendation Models: Progress and Directions. InCompanion Proceedings of the ACM on Web Conference 2025 (Sydney NSW, Australia)(WWW ’25). Association for Computi...

    2025

  2. [10]

    Wang-Cheng Kang and Julian McAuley. 2018. Self-Attentive Sequential Recom- mendation. arXiv:1808.09781 [cs.IR] https://arxiv.org/abs/1808.09781

    2018 arXiv

  3. [11]

    Yehuda Koren, Robert Bell, and Chris Volinsky. 2009. Matrix Factorization Techniques for Recommender Systems.Computer42, 8 (2009), 30–37. doi:10. 1109/MC.2009.263

    2009

  4. [12]

    Zhanyu Liu, Shiyao Wang, Xingmei Wang, Rongzhou Zhang, Jiaxin Deng, Honghui Bao, Jinghao Zhang, Wuchao Li, Pengfei Zheng, Xiangyu Wu, Yifei Hu, Qigen Hu, Xinchen Luo, Lejian Ren, Zixing Zhang, Qianqian Wang, Kuo Cai, Yunfan Wu, Hongtao Cheng, Zexuan Cheng, Lu Ren, Huanjie Wang...

    2025

  5. [13]

    Hanjia Lyu, Song Jiang, Hanqing Zeng, Yinglong Xia, Qifan Wang, Si Zhang, Ren Chen, Chris Leung, Jiajie Tang, and Jiebo Luo. 2024. LLM-Rec: Personalized Recommendation via Prompting Large Language Models. InFindings of the As- sociation for Computational Linguistics: NAACL 202...

    2024 doi

  6. [14]

    Jiaqi Ma, Zhe Zhao, Xinyang Yi, Jilin Chen, Lichan Hong, and Ed H. Chi. 2018. Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of- Experts(KDD ’18). Association for Computing Machinery, New York, NY, USA, 1930–1939. doi:10.1145/3219819.3220007

    2018

  7. [15]

    Maxim Naumov, Dheevatsa Mudigere, Hao-Jun Michael Shi, Jianyu Huang, Narayanan Sundaraman, Jongsoo Park, Xiaodong Wang, Udit Gupta, Carole-Jean Wu, Alisson G. Azzolini, Dmytro Dzhulgakov, Andrey Mallevich, Ilia Cherni- avskii, Yinghai Lu, Raghuraman Krishnamoorthi, Ansha Yu, V...

    2019 arXiv

  8. [16]

    Netflix Technology Blog. 2025. Foundation Model for Personalized Recom- mendation. https://netflixtechblog.com/foundation-model-for-personalized- recommendation-1a0bd8e02d39

    2025

  9. [17]

    Netflix Technology Blog. 2026. In-House LLM Serving at Netflix. https: //netflixtechblog.com/in-house-llm-serving-at-netflix-a5a8e799ea2c

    2026

  10. [18]

    Prithvi Rajasekaran, Ethan Dixon, Carly Ryan, and Jeremy Hadfield. 2025. Effec- tive context engineering for AI agents. https://www.anthropic.com/engineering/ effective-context-engineering-for-ai-agents. Anthropic Tech Blog

    2025

  11. [19]

    Tran, Jonah Samost, Maciej Kula, Ed H

    Shashank Rajput, Nikhil Mehta, Anima Singh, Raghunandan Keshavan, Trung Vu, Lukasz Heidt, Lichan Hong, Yi Tay, Vinh Q. Tran, Jonah Samost, Maciej Kula, Ed H. Chi, and Maheswaran Sathiamoorthy. 2023. Recommender systems with generative retrieval. InProceedings of the 37th Inter...

    2023

  12. [20]

    Anima Singh, Trung Vu, Nikhil Mehta, Raghunandan Keshavan, Maheswaran Sathiamoorthy, Yilin Zheng, Lichan Hong, Lukasz Heldt, Li Wei, Devansh Tandon, Ed Chi, and Xinyang Yi. 2024. Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations. InProceeding...

    2024 doi

  13. [21]

    Fei Sun, Jun Liu, Jian Wu, Changhua Pei, Xiao Lin, Wenwu Ou, and Peng Jiang

  14. [22]

    Gary Tang, Jiangwei Pan, Henry Wang, and Justin Basilico. 2023. Reward in- novation for long-term member satisfaction. InProceedings of the 17th ACM Conference on Recommender Systems(Singapore, Singapore)(RecSys ’23). Asso- ciation for Computing Machinery, New York, NY, USA, 3...

    2023

  15. [23]

    Likang Wu, Zhi Zheng, Zhaopeng Qiu, Hao Wang, Hongchao Gu, Tingjia Shen, Chuan Qin, Chen Zhu, Hengshu Zhu, Qi Liu, Hui Xiong, and Enhong Chen. 2024. A Survey on Large Language Models for Recommendation. arXiv:2305.19860 [cs.IR] https://arxiv.org/abs/2305.19860

    2024 arXiv

  16. [24]

    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. arXiv:2311.02089 [cs.IR] https://arxiv.org/abs/2311.02089

    2023 arXiv

  17. [25]

    Jiaqi Zhai, Lucy Liao, Xing Liu, Yueming Wang, Rui Li, Xuan Cao, Leon Gao, Zhao- jie Gong, Fangda Gu, Michael He, Yinghai Lu, and Yu Shi. 2024. Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations. arXiv:2402.17152 [cs.LG] h...

    2024 arXiv

  18. [26]

    Changsheng Zhao, Ernie Chang, Zechun Liu, Chia-Jung Chang, Wei Wen, Chen Lai, Sheng Cao, Yuandong Tian, Raghuraman Krishnamoorthi, Yangyang Shi, et al. 2025. Mobilellm-r1: Exploring the limits of sub-billion language model reasoners with open training recipes.arXiv preprint ar...

    2025

  19. [27]

    Zhe Zhao, Lichan Hong, Li Wei, Jilin Chen, Aniruddh Nath, Shawn Andrews, Aditee Kumthekar, Maheswaran Sathiamoorthy, Xinyang Yi, and Ed Chi. 2019. Recommending what video to watch next: a multitask ranking system. InPro- ceedings of the 13th ACM conference on recommender syste...

    2019

  20. [28]

    Guorui Zhou, Xiaoqiang Zhu, Chenru Song, Ying Fan, Han Zhu, Xiao Ma, Yanghui Yan, Junqi Jin, Han Li, and Kun Gai. 2018. Deep interest network for click-through rate prediction. InProceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining. ...

    2018

  21. [2016]

    Association for Computing Machinery, New York, NY, USA, 7–10

    Wide & Deep Learning for Recommender Systems(DLRS 2016). Association for Computing Machinery, New York, NY, USA, 7–10. doi:10.1145/2988450. 2988454 GenRec: An LLM-Backed Recommendation Ranker at Netflix

    2016 doi

  22. [2019]

    InProceedings of the 28th ACM International Conference on Information and Knowledge Management(Beijing, China)(CIKM ’19)

    BERT4Rec: Sequential Recommendation with Bidirectional Encoder Rep- resentations from Transformer. InProceedings of the 28th ACM International Conference on Information and Knowledge Management(Beijing, China)(CIKM ’19). Association for Computing Machinery, New York, NY, USA, ...

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.