Pith. sign in

REVIEW 2 major objections 5 minor 28 references

EGR: Embedding-Native Generative Retrieval with a Shared LLM

T0 review · 2 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read EGR claims that a single shared LLM, jointly trained on item-pair and history-to-target contrastive objectives, makes generative retrieval a plain ANN lookup and outperforms both semantic-ID and two-model generative retrieval.

desk verdict EGR is a clean, genuinely simpler retrieval design with good industrial results, but the 'shared LLM' benefit is not isolated by the main ablation. read the letter →

arxiv 2607.23038 v1 pith:6J46SWNW submitted 2026-07-25 cs.IR

Xiaodong Liu , Congfei Zhang , Hsiang-wei Chao , Siman Wang , Tong Zhao , Xiao Bai , Vincent Zhang , Jingxiao Ma
show 8 more authors
Zhe Liu Wenfeng Zhuo Zichu Li Jitin Krishnan Yunzhi Zhou Yajun Wang Jinchao Li Yu Zhang
This is my paper
classification cs.IR
keywords embedding-nativegenerativeretrievalsharedLLMjointcontrastivetrainingsemanticidentifiersapproximatenearestneighborsequentialrecommendationcold-startInfoNCE
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

EGR sets out to show that generative retrieval does not need discrete semantic identifiers or a separately trained, frozen item encoder. It uses one shared LLM backbone to embed item metadata into an indexed vector space and user interaction histories into dense queries, trained jointly with two contrastive losses: an item-pair loss that shapes the item space, and a history-to-target loss with a stop-gradient that teaches histories to retrieve future items. If the paper is right, the item space and the user-query space are one jointly learned coordinate system, and serving is just standard approximate-nearest-neighbor search. The authors support this with better recall on a public benchmark, monotonic gains as industrial training data grows, strong cold-start performance, a multimodal-input gain, and a live conversion-rate lift.

What carries the argument

The load-bearing object is the shared LLM backbone plus projection head, used in both paths, with a joint loss L = L_IRL + λ_t L_NIP. L_IRL is a symmetric contrastive (InfoNCE) loss over consecutive co-engaged item pairs; L_NIP is a history-to-target contrastive loss with a stop-gradient on the target item's embedding. The stop-gradient is the piece that lets item-space shaping and query-space training share one parameter set: NIP gradients flow through the history branch and shared parameters, helping future item embeddings, without letting per-example target pulls destabilize the indexed space. Work being done: the same parameters define both the coordinate system (items) and the map into

What would settle it

Run EGR with one change: detach the user-history path from the shared backbone (use a separate query encoder with identical architecture) while keeping the same joint loss and stop-gradient. If relative Recall@10 stays near the 100% level instead of falling toward the 86.2% two-stage level, then the shared-backbone coupling is not what drives the gains. A complementary check: measure how much item embeddings change after NIP is active versus an IRL-only run; if they are nearly identical, the NIP-to-item influence claimed through shared parameters is not happening.

Watch

Extended reading notes

Core claim

EGR's central discovery is that generative retrieval can be reduced to a single shared embedding space learned end-to-end. A pretrained vision-language model with low-rank adapters and a projection head encodes item metadata into dense vectors that go directly into an ANN index, and encodes a user's chronological engagement sequence into a dense query. The two objectives—symmetric item-pair contrastive learning and history-to-target contrastive learning with a stop-gradient on the target—are optimized jointly, so the indexed item space and the query side evolve together rather than being stitched in sequence. The paper claims this removes the quantization, mutable identifier vocabulary, and

Load-bearing premise

The load-bearing premise is that NIP gradients, flowing only through the history branch and shared parameters, can still reshape the indexed item space so it stays aligned with user queries—even though the NIP target item's embedding never receives a gradient directly. If that indirect coupling is weak, EGR's joint training effectively degenerates into a two-stage pipeline, which the paper's own ablation measures at only 86.2% relative Recall@10.

Editorial extensions

If this is right

  • Semantic-ID quantization, identifier vocabulary updates, and token-to-item grounding can be dropped entirely; the retriever outputs a plain dense vector.
  • Item space and query space are aligned during training, not after; the paper's ablation shows a two-stage pipeline (IRL then frozen, NIP on top) reaches only 86.2% of full EGR's relative Recall@10.
  • New or changed items can be added to the index by encoding their metadata, with no need to rebuild an ID vocabulary or retrain a generator.
  • Multimodal item input (text plus image) improves retrieval on the same backbone, suggesting richer metadata transfers directly into retrieval quality.
  • In a live production deployment with a fixed candidate quota, EGR as an added retrieval source lifted conversion rate by 2.91%, CTR by 0.23%, and impressions by 0.15%.

Reading between the lines

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

  • If the shared-backbone coupling is the active mechanism, the same two-loss joint-training recipe is a general pattern for any retrieval domain where item and query encoders are currently trained separately—this is our inference, not the paper's claim.
  • Because cold-start items are retrieved almost as well as head items, EGR-style metadata encoding could reduce the retraining cadence in rapidly churning catalogs; the paper documents the effect but does not quantify the operational cost savings.
  • A natural follow-up experiment the paper leaves implicit: vary the stop-gradient schedule (e.g., lift it in later training steps) to test whether the 100% configuration is the optimum or just a robust default.
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

2 major / 5 minor

Summary. The paper proposes EGR, an embedding-native generative retrieval framework in which a single shared LLM/MLLM backbone encodes item metadata into dense indexed vectors and user interaction histories into dense query vectors. The model is trained jointly with an item-pair InfoNCE loss (IRL) and a history-to-target InfoNCE loss (NIP) with a stop-gradient on the target item embedding, using a warm-up and ramp schedule for the NIP weight. Items are served through a standard ANN index, avoiding semantic-ID quantization and grounding. The paper evaluates EGR on Amazon Reviews, on Snap DPA data for scaling, cold-start, and multimodal behavior, and in a two-week online A/B test reporting a +2.91% CVR lift.

Significance. If the central architectural claim is sound, EGR would be a practically attractive design: it removes the quantization/grounding layer of SID-based retrieval and the frozen item encoder of the described two-model embedding pipelines, while preserving standard ANN serving. The industrial deployment, the honest cold-start caveat in §4.3.2, and the clear specification of the joint loss and stop-gradient mechanism are strengths. I also see no circularity: the reported results are empirical measurements, and the design choices are not fitted to the claimed outcomes. However, the key ablation in Table 4 does not isolate the shared-backbone mechanism from continued item-side training, and the public-benchmark comparison does not control for the large pretrained backbone used only by EGR. The significance of the architecture claim is therefore not yet established at the level the paper states.

major comments (2)
  1. [§4.4.1, Table 4] The 'Two models' row changes two variables at once relative to EGR: parameter sharing versus separate encoders, and trainable versus frozen item encoder. The 13.8-point gap (86.2% vs. 100.0%) is attributed in §3.4 to jointly shaping the item and query spaces through a shared backbone, but it could instead come from the item encoder continuing to receive IRL gradients while NIP trains. The §3.3 statement that NIP gradients 'can influence future item embeddings through the shared parameters' is an indirect effect whose strength is never measured; the 'w/o stop-grad' row (95.3%) removes a different mechanism and cannot serve as a control. A necessary experiment is a two-tower variant in which the item encoder is not frozen and keeps receiving IRL loss while a separate query model is trained with NIP. If that variant matches EGR, the shared-backbone claim is unsupported; if it does not, the
  2. [§4.1–4.2, Table 1] The Amazon Reviews comparison is not capacity-matched and has no uncertainty quantification. EGR initializes from Qwen3-VL-Embedding 2B, a large pretrained multimodal LLM with LoRA, while the compared baselines are trained from scratch or use item-ID embeddings. The margins are small on two of three datasets (Beauty 0.0655 vs. 0.0652; Sports 0.0362 vs. 0.0347), and no error bars, seeds, or significance tests are reported. This does not support the claim that EGR 'outperforms published baselines on Amazon Reviews' as stated. Please report multiple seeds with confidence intervals and add at least one same-backbone control, such as an SID-style or two-model embedding pipeline using the same Qwen3-VL-Embedding backbone and LoRA setup, to separate the effect of the pretrained encoder from the effect of the EGR training scheme.
minor comments (5)
  1. [§4.5, Table 6] The online A/B test compares a production stack with EGR added as an extra retrieval source against a production control; it is not an architecture-level comparison of EGR against a two-model or SID-based GR alternative. As such it supports system-level value but should be described more cautiously. Also, no confidence interval or statistical significance is reported for the +2.91% CVR lift.
  2. [§4.1, Implementation] No code or data release is described, and Table 1 numbers appear to be single runs. Given the reliance on an external backbone and LoRA hyperparameters, reporting seeds and model-card versions would substantially improve reproducibility.
  3. [§2, Related Work] The statement that embedding-based pipelines 'train the item encoder separately from the query generator' is presented as a general property, but the cited systems (PinRec, TBGRecall) are not shown to freeze the item encoder. The contrast in §3.4 would be cleaner if the paper distinguished the authors' design assumption from an observed property of all cited works.
  4. [§3.4, Joint Training] The warm-up and λ_t schedule is described only in prose. Writing it as an explicit equation, e.g., λ_t = 0 for the first 10% of steps and then a linear ramp from 0.1 to 1.0, would remove ambiguity about when the ramp begins.
  5. [§4.3.1, Figure 3] With only four training-size points, 'no observed saturation' is stronger than warranted. A statement such as 'improvement continued over the range tested' would better match the evidence.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: EGR's claims are empirical evaluations, not derivations that reduce to their inputs.

full rationale

The paper's central claim is that a shared LLM backbone jointly trained with IRL and NIP improves retrieval over SID-based and two-model embedding-based generative retrieval. This claim is supported by held-out Recall@10 measurements on Amazon Reviews and Snap DPA, an offline cold-start frequency analysis, multimodal comparisons, and a live two-week A/B test with a +2.91% CVR lift. None of these results are derived by construction from the loss definitions or from fitted parameters renamed as predictions. The NIP objective (Eq. 4) directly optimizes the query embedding to align with the target item embedding, but that is a standard supervised contrastive formulation, not a circular prediction: the target item embedding is computed from held-out metadata and the evaluation measures retrieval on unseen sequences. The stop-gradient and loss-schedule ablations are design choices tested empirically, not parameters fitted to the evaluation metric. The paper cites prior work and the Qwen3-VL-Embedding backbone, but its central architecture and results do not depend on any self-citation chain or imported uniqueness theorem. The strongest skeptical concern—that the shared-backbone benefit is confounded with the trainable item encoder—is an experimental-design limitation about causal attribution, not circularity. The derivation chain is therefore self-contained with respect to circularity, and the appropriate score is 0.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

EGR introduces no new physical or latent entities; its components are standard encoders, projection heads, InfoNCE losses, and prompts. The free parameters are empirical hyperparameters rather than fitted scientific constants; the axioms are the domain assumptions on which the empirical method rests.

free parameters (5)
  • InfoNCE temperature τ
    Learned temperature in Eq. (2)/(4); controls contrastive sharpness and is optimized during training.
  • NIP loss weight λ_t schedule = 0 (first 10%), ramps 0.1→1.0
    Hand-chosen schedule; ablations compare Joint/Balanced/Sequential but no principled derivation.
  • Projection head output dim = 512
    L2-normalized embedding dim chosen for compact ANN serving; no sensitivity analysis.
  • LoRA rank/alpha = r=128, α=256
    Chosen adapter configuration (~41M trainable params); not derived.
  • Training hyperparameters = lr=2e-5, batch=64/GPU, max history=20, one epoch
    Chosen recipe; affects results and not determined by theory.
assumptions (5)
  • domain assumption Consecutive co-engagement pairs define useful item similarity for retrieval
    IRL's entire supervision is that next items are similar; if co-engagement structure does not align with future-item retrieval, the item-space foundation is wrong (§3.2, Eq. 2).
  • domain assumption A single LLM + MLP can map item metadata and user histories into one comparable L2-normalized space
    Shared backbone assumed to produce compatible item/query vectors for cosine/MIP ANN search (§3.1, Eq. 1/3).
  • domain assumption Stop-gradient on NIP target still yields joint alignment through shared parameters
    The paper asserts this rather than proves it; central to the joint-training claim (§3.3).
  • domain assumption In-batch negatives across DDP ranks are sufficient for item discrimination at scale
    No hard-negative mining or large-corpus negative sampling is discussed; contrastive denominator is limited to batch size × GPUs (§4.1).
  • domain assumption Qwen3-VL-Embedding 2B's pretrained priors transfer to item metadata and cold-start products
    Cold-start/multimodal gains rely on pretrained backbone rather than EGR-specific machinery (§4.3.2, §4.3.3).

how reviews work

0 comments
Cite this review

Pith. "Pith review of EGR: Embedding-Native Generative Retrieval with a Shared LLM." pith.science (2026). https://pith.science/paper/6J46SWNW

@misc{pith2026260723038,
  author       = {Pith},
  title        = {Pith review of: EGR: Embedding-Native Generative Retrieval with a Shared LLM},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6J46SWNW}},
  note         = {Machine review of arXiv:2607.23038}
}
read the original abstract

Generative retrieval is increasingly popular in large-scale recommendation and advertising systems, yet current methods introduce practical complications. Semantic-ID methods rely on quantization, mutable identifier vocabularies, and token-to-item grounding; embedding-based pipelines train the item encoder separately from the query generator, which limits user-item alignment. We propose EGR, an Embedding-Native Generative Retrieval framework for recommendation and advertising. EGR uses a single shared LLM to learn item representations from item metadata and user representations from interaction histories in one embedding space. Items are indexed directly as dense vectors, and user histories are encoded as dense retrieval queries. Joint contrastive training groups related items and aligns queries with their target items. We evaluate EGR on public benchmarks, industrial data, and live deployment. EGR outperforms published baselines on Amazon Reviews; on Snap DPA, it scales with data, handles cold-start items, and benefits from multimodal input. In production, EGR delivers a +2.91% conversion-rate lift, simplifying system design while improving retrieval quality and ad performance.

Figures

Figures reproduced from arXiv: 2607.23038 by the authors.

Figure 1
Figure 1. Comparison of generative retrieval families. (a) [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overall framework of EGR. A shared LLM with a projection head encodes both item and user histories. IRL applies a [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Data scaling of EGR. R@10 normalized to its value [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

28 extracted references · 4 linked inside Pith

  1. [1]

    Prabhat Agarwal, Anirudhan Badrinath, Laksh Bhasin, Jaewon Yang, Edoardo Botta, Jiajing Xu, and Charles Rosenberg. 2025. PinRec: Outcome-Conditioned, Multi-Token Generative Retrieval for Industry-Scale Recommendation Systems. arXiv:2504.10507 [cs.IR] https://arxiv.org/abs/2504.10507

    arXiv 2025

  2. [2]

    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

  3. [3]

    Paul Covington, Jay Adams, and Emre Sargin. 2016. Deep Neural Networks for YouTube Recommendations. InProceedings of the 10th ACM Conference on Recommender Systems (RecSys ’16). ACM, New York, NY, USA, 191–198. doi:10. 1145/2959100.2959190

    arXiv 2016

  4. [4]

    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

  5. [5]

    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. 2025. PLUM: Adapting Pre-trained Langua...

    arXiv 2025

  6. [6]

    Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen

    Edward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. LoRA: Low-Rank Adap- tation of Large Language Models. InThe Tenth International Conference on Learning Representations (ICLR 2022). OpenReview.net, Virtual Event, 13 pages. https://openreview.net/forum?id=nZeVKeeFYf9

    2022

  7. [7]

    Wang-Cheng Kang and Julian McAuley. 2018. Self-Attentive Sequential Recom- mendation. In2018 IEEE International Conference on Data Mining (ICDM). IEEE, Singapore, 197–206. doi:10.1109/ICDM.2018.00035

    arXiv 2018

  8. [8]

    Zida Liang, Changfa Wu, Dunxian Huang, Weiqiang Sun, Ziyang Wang, Yuliang Yan, Jian Wu, Yuning Jiang, Bo Zheng, Ke Chen, Silu Zhou, and Yu Zhang. 2025. EGR TBGRecall: A Generative Retrieval Model for E-commerce Recommendation Scenarios. arXiv:2508.11977 [cs.IR] https://arxiv.org/abs/2508.11977

    arXiv 2025

Show all 28 references
  1. [9]

    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

  2. [10]

    Julian McAuley, Christopher Targett, Qinfeng Shi, and Anton van den Hengel

  3. [11]

    Gustavo Penha, Edoardo D’Amico, Marco De Nadai, Enrico Palumbo, Alexandre Tamborrino, Ali Vardasbi, Max Lefarov, Shawn Lin, Timothy Heath, Francesco Fabbri, and Hugues Bouchard. 2025. Semantic IDs for Joint Generative Search and Recommendation. InProceedings of the Nineteenth ...

    2025

  4. [12]

    Qwen Team. 2025. Qwen3-VL-Embedding-2B. Hugging Face model card. https: //huggingface.co/Qwen/Qwen3-VL-Embedding-2B

    2025

  5. [13]

    Keshavan, Trung Vu, Lukasz Heldt, Lichan Hong, Yi Tay, Vinh Q

    Shashank Rajput, Nikhil Mehta, Anima Singh, Raghunandan H. Keshavan, Trung Vu, Lukasz Heldt, Lichan Hong, Yi Tay, Vinh Q. Tran, Jonah Samost, Maciej Kula, Ed H. Chi, and Maheswaran Sathiamoorthy. 2023. Recommender Systems with Generative Retrieval. arXiv:2305.05065 [cs.IR] htt...

    2023 arXiv

  6. [14]

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

  7. [15]

    Juntao Tan, Shuyuan Xu, Wenyue Hua, Yingqiang Ge, Zelong Li, and Yongfeng Zhang. 2024. IDGenRec: LLM-RecSys Alignment with Textual ID Learning. InProceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR ’24). ACM, N...

    2024

  8. [16]

    Aäron van den Oord, Yazhe Li, and Oriol Vinyals. 2018. Representation Learning with Contrastive Predictive Coding. arXiv:1807.03748 [cs.LG] https://arxiv.org/ abs/1807.03748

    2018 arXiv

  9. [17]

    Huanjie Wang, Xinchen Luo, Honghui Bao, Zixing Zhang, Lejian Ren, Yun- fan Wu, Hongwei Zhang, Liwei Guan, and Guang Chen. 2026. PIT: A Dy- namic Personalized Item Tokenizer for End-to-End Generative Recommendation. arXiv:2602.08530 [cs.IR] https://arxiv.org/abs/2602.08530

    2026

  10. [18]

    Ben Xue, Dan Liu, Lixiang Wang, Mingjie Sun, Peng Wang, Pengfei Zhang, Shaoyun Shi, Tianyu Xu, Yunhao Sha, Zhiqiang Liu, Bo Kong, Bo Wang, Hang Yang, Jieting Xue, Junhao Wang, Shengyu Wang, Shuping Hui, Wencai Ye, Xiao Lin, Yongzhi Li, Yuhang Chen, Zhihui Yin, Quan Chen, Shiya...

    2026

  11. [19]

    Wencai Ye, Mingjie Sun, Shaoyun Shi, Peng Wang, Wenjin Wu, and Peng Jiang

  12. [20]

    Xinyang Yi, Ji Yang, Lichan Hong, Derek Zhiyuan Cheng, Lukasz Heldt, Aditee Kumthekar, Zhe Zhao, Li Wei, and Ed H. Chi. 2019. Sampling-Bias-Corrected Neural Modeling for Large Corpus Item Recommendations. InProceedings of the 13th ACM Conference on Recommender Systems (RecSys ...

    2019

  13. [21]

    Hamilton, and Jure Leskovec

    Rex Ying, Ruining He, Kaifeng Chen, Pong Eksombatchai, William L. Hamilton, and Jure Leskovec. 2018. Graph Convolutional Neural Networks for Web-Scale Recommender Systems. InProceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD ’...

    2018

  14. [22]

    Jiaqi Zhai, Lucy Liao, Xing Liu, Yueming Wang, Rui Li, Xuan Cao, Leon Gao, Zhaojie Gong, Fangda Gu, Jiayuan He, Yinghai Lu, and Yu Shi. 2024. Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Gener- ative Recommendations. InProceedings of the 41st ...

    2024

  15. [23]

    Jun Zhang, Yi Li, Yue Liu, Changping Wang, Yuan Wang, Yuling Xiong, Xun Liu, Haiyang Wu, Qian Li, Enming Zhang, Jiawei Sun, Xin Xu, Zishuai Zhang, Ruoran Liu, Suyuan Huang, Zhaoxin Zhang, Zhengkai Guo, Shuojin Yang, Meng-Hao Guo, Huan Yu, Jie Jiang, and Shi-Min Hu. 2025. GPR: ...

    2025

  16. [24]

    Zhiyang Zhang, Junda She, Kuo Cai, Bo Chen, Shiyao Wang, Xinchen Luo, Qiang Luo, Ruiming Tang, Han Li, Kun Gai, and Guorui Zhou. 2026. Unleashing the Native Recommendation Potential: LLM-Based Generative Recommendation via Structured Term Identifiers. arXiv:2601.06798 [cs.IR] ...

    2026

  17. [2015]

    InProceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)

    Image-Based Recommendations on Styles and Substitutes. InProceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR). Association for Computing Machinery, New York, NY, USA, 43–52. doi:10.1145/2766462.2767755

  18. [2016]

    InProceedings of the 1st Workshop on Deep Learning for Recommender Systems (DLRS 2016)

    Wide & Deep Learning for Recommender Systems. InProceedings of the 1st Workshop on Deep Learning for Recommender Systems (DLRS 2016). ACM, New York, NY, USA, 7–10. doi:10.1145/2988450.2988454

    2016

  19. [2019]

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

    BERT4Rec: Sequential Recommendation with Bidirectional Encoder Rep- resentations from Transformer. InProceedings of the 28th ACM International Conference on Information and Knowledge Management (CIKM ’19). ACM, New York, NY, USA, 1441–1450. doi:10.1145/3357384.3357895

  20. [2025]

    arXiv:2508.10584 [cs.IR] https://arxiv.org/abs/2508.10584

    DAS: Dual-Aligned Semantic IDs Empowered Industrial Recommender System. arXiv:2508.10584 [cs.IR] https://arxiv.org/abs/2508.10584

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.