Pith. sign in

REVIEW 3 major objections 5 minor 41 references

Knowledge Distillation for Enhancing Walmart E-commerce Search Relevance Using Large Language Models

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read The paper claims that a BERT-base student distilled from a 7B LLM teacher matches or slightly exceeds it on ranking metrics when trained on 170M teacher-labeled query-item pairs.

desk verdict Solid industrial distillation scaling study; the margin-MSE-on-unlabeled-data extension is useful, but the student-vs-teacher comparison undercuts itself by not ruling out golden-test overlap with teacher-labeled training data. read the letter →

arxiv 2505.07105 v1 pith:BLRNBNQ4 submitted 2025-05-11 cs.IR cs.LG

Hongwei Shang , Nguyen Vo , Nitin Yadav , Tian Zhang , Ajit Puthenputhussery , Xunfan Cai , Shuyi Chen , Prijith Chandran
show 1 more author
Changsung Kang
This is my paper · ORCID
classification cs.IRcs.LG
keywords LLMknowledgedistillatione-commercesearchMargin-MSElosscross-encoderrelevancetailqueriesproductiondeployment
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 claims that a small, deployable search-relevance model can absorb the ranking ability of a 7-billion-parameter LLM through knowledge distillation, provided the student is trained on a much larger set of teacher-labeled data than the teacher itself used. The student is a BERT-base cross-encoder with 110M parameters, trained with a Margin-MSE loss that matches the predicted relevance gaps between pairs of products for a query. On a golden tail-query test set, the student reaches roughly the same gains over the production baseline as the Mistral-7B teacher on classification-oriented recall, and slightly larger gains on NDCG. The paper reports that the student was deployed in production and improved tail-query engagement metrics such as add-to-cart rate per visitor (+1.2%, p<0.05).

What carries the argument

The mechanism is an extension of Margin-MSE distillation. For each query $q$, the teacher produces scores $t(q,d)$ and the student produces scores $s(q,d)$ for documents $d$; the loss averages the squared difference $(\Delta^t_{q,d_i,d_j} - \Delta^s_{q,d_i,d_j})^2$ over all unordered document pairs, so the student learns the teacher's relative gaps between products rather than absolute scores. This margin matching removes the need for true labels, which is what lets the framework convert large-scale unlabeled search-log data into 50M, 110M, and 170M teacher-labeled query-item pairs for student training.

What would settle it

Inspect the 2,354 golden test queries against the 5.2M unique queries in the llm-judged datasets. If a meaningful fraction of test queries appear in the teacher-labeled training data, retrain the student on data with those queries removed and re-measure the NDCG@5 lift over the baseline; the central claim predicts the lift will remain near +1.5%, while a large drop would indicate leakage-driven inflation.

Watch

Extended reading notes

Core claim

The central claim is that a BERT-base student model, distilled from a Mistral-7B teacher on 170M teacher-labeled query-item pairs, performs comparably to or slightly better than the teacher on ranking metrics, despite being more than 60 times smaller and using fewer input fields. Teacher and student share a cross-encoder architecture, but the student is trained on an expanded unlabeled dataset labeled entirely by the teacher, with the loss defined over all product-pair score margins for each query. The authors interpret the results as evidence that larger model capacity and the item-description field are not necessary to reproduce the teacher's relevance judgments, and that scaling teacher-labeled unlabeled data is the main driver of the student's gains.

Load-bearing premise

The load-bearing assumption is that the golden test queries and the teacher-labeled training pairs come from disjoint search traffic; the paper does not state that the 2,354 test queries were excluded from the llm-judged datasets, so if they overlap, part of the student's apparent parity with the teacher could be memorization of teacher scores rather than learned generalization.

Editorial extensions

If this is right

  • A 110M-parameter student can replace a 7B LLM in latency-sensitive ranking, making LLM-level relevance feasible in real-time e-commerce search.
  • Increasing teacher-labeled unlabeled data continues to improve student ranking quality, so more search-log data can be converted into training signal without human annotation.
  • Learning margins between pairs is more effective than pointwise score matching even when student and teacher share the same architecture.
  • The student's comparable performance without item descriptions means production inputs can be shortened while retaining teacher-level relevance.
  • Deploying the distilled student can yield measurable engagement gains on tail queries, including higher add-to-cart rates and lower session abandonment.

Reading between the lines

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

  • An extension the paper leaves implicit: the same data-generation loop could be applied to head and torso queries, where the paper's stated future work points, since the limiting factor appears to be labeled-data volume rather than model capacity.
  • A testable extension: because the all-pairs margin loss scales quadratically with the number of documents per query, sampling more than the reported 10 candidates per query may change how quickly performance plateaus.
  • A risk the paper does not discuss: training on 170M teacher-labeled pairs copies whatever bias the teacher has, such as overconfidence on popular brands, so calibrating teacher scores before labeling is a natural next check.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. This paper proposes a knowledge-distillation framework for e-commerce search relevance at Walmart. A 7B-parameter LLM teacher (Llama2-7B or Mistral-7B) is fine-tuned with soft relevance labels on human editorial data, then used to label large-scale unlabeled query-item pairs extracted from search logs. A BERT-base cross-encoder student is trained with a Margin-MSE loss on up to 170M such teacher-labeled pairs. Offline experiments on a golden tail-query test set show that Margin-MSE outperforms pointwise cross-entropy, that increasing teacher-labeled data improves NDCG, and that the student approaches or slightly exceeds the teacher on NDCG despite being much smaller. The student (XE v2) was deployed as a ranking feature, with significant lifts in manual evaluation, interleaving ATC, and an online AB test.

Significance. If validated, this is a practically significant industrial contribution: it demonstrates a concrete path to deploying LLM-level relevance judgments in a latency-constrained reranking system, with production-scale evidence (offline golden test plus online AB/interleaving). The paper's strengths are its clear ablation isolating loss function and data scale, and its honest reporting of deployment results. However, the headline comparison to the teacher is currently supported only by small offline metric differences without confidence intervals, and two methodological controls (golden/training overlap and teacher training-data asymmetry) must be resolved before the central claims can be taken at face value.

major comments (3)
  1. [§4.1 (Test Data)] The statement that the golden test set 'was generated after all human evaluation data ... was collected' addresses only human-judged datasets; it does not address the llm-judged_v1/v2/v3 training sets, which are sampled from the same tail search traffic (Table 3). Because llm-judged_v3 contains 5.2M unique queries, exact query-level overlap with the 2,354 golden queries is plausible. If any golden query appears in llm-judged_v3, XE v2 is trained with Margin-MSE on teacher scores for the very query-item pairs used for evaluation, and the reported NDCG@5 lift (+1.5% vs. teacher +1.33%) could reflect memorization rather than generalization. Please report overlap statistics or evaluate on a provably disjoint subset.
  2. [§4.2.1 and Table 5] The RQ3 comparison between the student and the teacher is confounded by training-data asymmetry. The Mistral-7B teacher is fine-tuned only on human-judged_v1 (6.0M QIPs), while XE v2 and XE v1.4 are initialized from XE v1.1, which was trained on human-judged_v2 (10M QIPs). Thus the student has access to more human-annotated data than the teacher in addition to the llm-judged data, so the apparent NDCG advantage over the teacher cannot be attributed solely to distillation at scale. To support the abstract's claim that 'with enough augmented data the student can outperform the teacher,' either train the teacher on human-judged_v2 as well, or initialize the student from a checkpoint trained only on human-judged_v1, or add an ablation that isolates this factor.
  3. [§4.3.3 and Tables 4-5] The offline metrics are reported without confidence intervals or significance tests, and the differences that support the 'student outperforms teacher' claim are small (NDCG@5 +1.50% vs +1.33%; R@P=95% +12.84% vs +13.27%). These gaps are within the range where sampling noise could change the qualitative conclusion. Please report bootstrap confidence intervals over the 2,354 golden queries or paired significance tests for the comparisons in RQ1-RQ3.
minor comments (5)
  1. [§5.3] The text refers to 'Table 8' for the AB test results, but the table is numbered Table 7.
  2. [§5.3/Table 7] The table marks all four metrics as statistically significant, but p-values are only given in the text for ATC rate per visitor; please report p-values or confidence intervals for all four metrics.
  3. [§3.2] The phrase 'softRank Adaptation (LoRA)' should read 'Low-Rank Adaptation (LoRA)'.
  4. [§4.3/Tables 4-5] The baseline rows are all listed as 0%, which likely means the reported numbers are relative lifts over XE v1; please state this explicitly and include the absolute baseline values for the golden test set.
  5. [§1] There is a typo in 'we present our our enhanced'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the student model is evaluated on an external human-labeled golden set and online AB metrics; self-citations are implementation-level and not load-bearing.

full rationale

The derivation chain is not circular. The Mistral-7B teacher is fine-tuned on the human-judged_v1 dataset with a binary soft-target cross-entropy loss, and the XE v2 student is trained on teacher-generated labels on the llm-judged_v3 dataset with an extended Margin-MSE loss. The central claim that XE v2 approaches or exceeds the teacher is measured on a separately human-labeled golden test set of 2,354 tail queries and 32,586 QIPs, plus online interleaving and AB tests, not on the teacher's own training labels. The student's training objective is to match teacher score margins on augmented unlabeled data; it is not constructed to match the teacher's scores on the golden queries, so the reported NDCG@5, R@P, and ATC lifts are not equal by construction to any fitted input. Self-citations to the authors' prior work [18] and [31] are used for implementation choices such as LoRA rank, soft-target formulation, the Margin-MSE loss, and the benefit of item descriptions; these citations are not invoked as the load-bearing evidence for the data-scaling result, nor do they forbid alternative approaches. The only substantive validity concern is whether tail queries in the golden test set overlap with queries in llm-judged_v3; Section 4.1's no-leakage statement addresses human evaluation data only. That is a data-hygiene or leakage risk, not a circularity, because even if overlap existed, the student's human-labeled NDCG would not be forced by the teacher-margin training objective.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim rests on standard supervised learning assumptions plus domain-specific choices: the soft-label mapping and the proxy quality of the teacher. No new entities are introduced. The main free parameters are hyperparameters borrowed from prior work, and the key assumption is that teacher-generated labels on unlabeled logs are a valid training signal.

free parameters (5)
  • Soft label mapping = ratings 4->1, 3->0.5, 0-2->0
    Design choice converting 5-scale editorial labels to soft targets; directly affects the teacher training objective.
  • LoRA rank = 256
    Chosen based on prior work [18]; not independently validated in this paper.
  • LoRA alpha = 128
    Hyperparameter from [18], used for teacher fine-tuning.
  • Learning rate = 1e-5
    Selected from {1e-4, 1e-5} on the teacher; same value used for the student.
  • Dropout = 0.05
    Set from prior work [18]; not tuned in this paper.
assumptions (4)
  • domain assumption Human editorial ratings are the ground truth for relevance.
    The teacher is fine-tuned on human labels and the golden test set uses human ratings; all evaluation depends on this.
  • domain assumption The golden test set is representative of tail queries and its offline metrics align with online human evaluations.
    Section 4.1 asserts this alignment to justify the small test set; it is not independently validated here.
  • domain assumption Margin-MSE is an effective distillation objective for same-architecture cross-encoders.
    Adopted from Hofstätter et al. [7]; central to the student training and validated only by the ablation in this paper.
  • domain assumption The teacher's predictions on unlabeled search-log data provide useful supervision for the student.
    This is the core of the augmented-data approach; the paper assumes teacher scores are a reliable learning signal.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Knowledge Distillation for Enhancing Walmart E-commerce Search Relevance Using Large Language Models." pith.science (2026). https://pith.science/paper/BLRNBNQ4

@misc{pith2026250507105,
  author       = {Pith},
  title        = {Pith review of: Knowledge Distillation for Enhancing Walmart E-commerce Search Relevance Using Large Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BLRNBNQ4}},
  note         = {Machine review of arXiv:2505.07105}
}
read the original abstract

Ensuring the products displayed in e-commerce search results are relevant to users queries is crucial for improving the user experience. With their advanced semantic understanding, deep learning models have been widely used for relevance matching in search tasks. While large language models (LLMs) offer superior ranking capabilities, it is challenging to deploy LLMs in real-time systems due to the high-latency requirements. To leverage the ranking power of LLMs while meeting the low-latency demands of production systems, we propose a novel framework that distills a high performing LLM into a more efficient, low-latency student model. To help the student model learn more effectively from the teacher model, we first train the teacher LLM as a classification model with soft targets. Then, we train the student model to capture the relevance margin between pairs of products for a given query using mean squared error loss. Instead of using the same training data as the teacher model, we significantly expand the student model dataset by generating unlabeled data and labeling it with the teacher model predictions. Experimental results show that the student model performance continues to improve as the size of the augmented training data increases. In fact, with enough augmented data, the student model can outperform the teacher model. The student model has been successfully deployed in production at Walmart.com with significantly positive metrics.

Figures

Figures reproduced from arXiv: 2505.07105 by the authors.

Figure 1
Figure 1. The architecture for the previous production ver [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Our proposed knowledge distillation framework [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 21 canonical work pages

  1. [1]

    Xiao Bai, Lei Duan, Richard Tang, Gaurav Batra, and Ritesh Agrawal. 2022. Improving text-based similar product recommendation for dynamic product advertising at yahoo. In Proceedings of the 31st ACM International Conference on Information & Knowledge Management . 2883–2892

    work page 2022

  2. [2]

    Scott Deerwester, Susan T Dumais, George W Furnas, Thomas K Landauer, and Richard Harshman. 1990. Indexing by latent semantic analysis. Journal of the American society for information science 41, 6 (1990), 391–407

    1990

  3. [3]

    Philipp Hager, Romain Deffayet, Jean-Michel Renders, Onno Zoeter, and Maarten de Rijke. 2024. Unbiased Learning to Rank Meets Reality: Lessons from Baidu’s Large-Scale Search Dataset. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval . 1546–1556

    work page 2024

  4. [4]

    Xingwei He, Yeyun Gong, A Jin, Weizhen Qi, Hang Zhang, Jian Jiao, Bartuer Zhou, Biao Cheng, Siu-Ming Yiu, Nan Duan, et al . 2022. Metric-guided distillation: Distilling knowledge from the metric to ranker and retriever for generative commonsense reasoning. arXiv preprint arXiv:2210.11708 (2022)

    work page Pith review arXiv 2022

  5. [5]

    Yunzhong He, Yuxin Tian, Mengjiao Wang, Feier Chen, Licheng Yu, Mao- long Tang, Congcong Chen, Ning Zhang, Bin Kuang, and Arul Prakash. 2023. Que2engage: Embedding-based retrieval for relevant and engaging products at facebook marketplace. In Companion Proceedings of the ACM Web Conference

    2023

  6. [6]

    Geoffrey Hinton. 2015. Distilling the Knowledge in a Neural Network. arXiv preprint arXiv:1503.02531 (2015)

    arXiv 2015

  7. [7]

    Sebastian Hofstätter, Sophia Althammer, Michael Schröder, Mete Sertkan, and Allan Hanbury. 2020. Improving efficient neural ranking models with cross- architecture knowledge distillation. arXiv preprint arXiv:2010.02666 (2020)

    arXiv 2020

  8. [8]

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. Lora: Low-rank adaptation of large language models. International Conference on Learning Representations (2022)

    work page 2022

Show all 41 references
  1. [9]

    Albert Q Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, De- vendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, et al . 2023. Mistral 7B. arXiv preprint arXiv:2310.06825 (2023)

    2023 arXiv

  2. [10]

    Thorsten Joachims et al. 2003. Evaluating Retrieval Performance Using Click- through Data

    2003

  3. [11]

    Jacob Devlin Ming-Wei Chang Kenton and Lee Kristina Toutanova. 2019. Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of naacL-HLT, Vol. 1. Minneapolis, Minnesota, 2

    2019

  4. [12]

    Juexin Lin, Sachin Yadav, Feng Liu, Nicholas Rossi, Praveen R Suram, Satya Chem- bolu, Prijith Chandran, Hrushikesh Mohapatra, Tony Lee, Alessandro Magnani, et al. 2024. Enhancing Relevance of Embedding-based Retrieval at Walmart. In Proceedings of the 33rd ACM International C...

    2024

  5. [13]

    Shilei Liu, Lin Li, Jun Song, Yonghua Yang, and Xiaoyi Zeng. 2023. Multimodal pre-training with self-distillation for product understanding in e-commerce. In Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining. 1039–1047

    2023

  6. [14]

    Yufan Liu, Jiajiong Cao, Bing Li, Chunfeng Yuan, Weiming Hu, Yangxi Li, and Yunqiang Duan. 2019. Knowledge distillation via instance relationship graph. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 7096–7104

    2019

  7. [15]

    Ziyang Liu, Chaokun Wang, Hao Feng, Lingfei Wu, and Liqun Yang. 2022. Knowl- edge distillation based contextual relevance matching for e-commerce product search. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: Industry Track . 63–76

    2022

  8. [16]

    Chen Luo, Xianfeng Tang, Hanqing Lu, Yaochen Xie, Hui Liu, Zhenwei Dai, Limeng Cui, Ashutosh Joshi, Sreyashi Nag, Yang Li, et al. 2024. Exploring Query Understanding for Amazon Product Search. arXiv preprint arXiv:2408.02215 (2024)

    2024 arXiv

  9. [17]

    Alessandro Magnani, Feng Liu, Suthee Chaidaroon, Sachin Yadav, Praveen Reddy Suram, Ajit Puthenputhussery, Sijie Chen, Min Xie, Anirudh Kashi, Tony Lee, et al. 2022. Semantic Retrieval at Walmart. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data...

    2022

  10. [18]

    Navid Mehrdad, Hrushikesh Mohapatra, Mossaab Bagdouri, Prijith Chandran, Alessandro Magnani, Xunfan Cai, Ajit Puthenputhussery, Sachin Yadav, Tony Lee, ChengXiang Zhai, et al. 2024. Large Language Models for Relevance Judgment in Product Search. arXiv preprint arXiv:2406.00247 (2024)

    2024 arXiv

  11. [19]

    Aditya Menon, Sadeep Jayasumana, Ankit Singh Rawat, Seungyeon Kim, Sashank Reddi, and Sanjiv Kumar. 2022. In defense of dual-encoders for neural ranking. In International Conference on Machine Learning . PMLR, 15376–15400

    2022

  12. [20]

    Priyanka Nigam, Yiwei Song, Vijai Mohan, Vihan Lakshman, Weitian Ding, Ankit Shingavi, Choon Hui Teo, Hao Gu, and Bing Yin. 2019. Semantic product search. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining . 2876–2885

    2019

  13. [21]

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. 2019. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing system...

    2019

  14. [22]

    Wenjun Peng, Guiyang Li, Yue Jiang, Zilong Wang, Dan Ou, Xiaoyi Zeng, Derong Xu, Tong Xu, and Enhong Chen. 2024. Large language model based long-tail query rewriting in taobao search. In Companion Proceedings of the ACM on Web Conference 2024. 20–28

    2024

  15. [23]

    Zhiyuan Peng, Vachik Dave, Nicole McNabb, Rahul Sharnagat, Alessandro Mag- nani, Ciya Liao, Yi Fang, and Sravanthi Rajanala. 2023. Entity-aware Multi-task Learning for Query Understanding at Walmart. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and D...

    2023

  16. [24]

    Stephen Robertson, Hugo Zaragoza, et al . 2009. The probabilistic relevance framework: BM25 and beyond. Foundations and Trends® in Information Retrieval 3, 4 (2009), 333–389

    2009

  17. [25]

    Weiwei Sun, Zheng Chen, Xinyu Ma, Lingyong Yan, Shuaiqiang Wang, Pengjie Ren, Zhumin Chen, Dawei Yin, and Zhaochun Ren. 2023. Instruction distilla- tion makes large language models efficient zero-shot rankers. arXiv preprint arXiv:2311.01555 (2023)

    2023 arXiv

  18. [26]

    Xing Tan, Fanghong Jiang, and Jimmy Xiangji Huang. 2018. StatBM25: An Aggregative and Statistical Approach for Document Ranking. In Proceedings of the 2018 ACM SIGIR International Conference on Theory of Information Retrieval . 207–210

    2018

  19. [27]

    Paul Thomas, Seth Spielman, Nick Craswell, and Bhaskar Mitra. 2024. Large language models can accurately predict searcher preferences. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1930–1940. Knowledge Disti...

    2024

  20. [28]

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

    2023 arXiv

  21. [29]

    Andrew Trotman, Antti Puurula, and Blake Burgess. 2014. Improvements to BM25 and language models examined. In Proceedings of the 2014 Australasian Document Computing Symposium. 58–65

    2014

  22. [30]

    Han Vanholder. 2016. Efficient inference with tensorrt. In GPU Technology Conference, Vol. 1

    2016

  23. [31]

    Nguyen Vo, Hongwei Shang, Zhen Yang, Juexin Lin, SD Mohseni Taheri, and Changsung Kang. 2024. Knowledge Distillation for Efficient and Effective Rele- vance Search on E-commerce. SIGIR eCom (2024)

    2024

  24. [32]

    Binbin Wang, Mingming Li, Zhixiong Zeng, Jingwei Zhuo, Songlin Wang, Sulong Xu, Bo Long, and Weipeng Yan. 2023. Learning Multi-Stage Multi-Grained Semantic Embeddings for E-Commerce Search. In Companion Proceedings of the ACM Web Conference 2023. 411–415

    2023

  25. [33]

    Han Wang, Mukuntha Narayanan Sundararaman, Onur Gungor, Yu Xu, Krishna Kamath, Rakesh Chalasani, Kurchi Subhra Hazra, and Jinfeng Rao. 2024. Improv- ing Pinterest Search Relevance Using Large Language Models. arXiv preprint arXiv:2410.17152 (2024)

    2024 arXiv

  26. [34]

    T Wolf. 2019. Huggingface’s transformers: State-of-the-art natural language processing. arXiv preprint arXiv:1910.03771 (2019)

    2019 arXiv

  27. [35]

    Xuyang Wu, Ajit Puthenputhussery, Hongwei Shang, Changsung Kang, and Yi Fang. 2024. Meta Learning to Rank for Sparsely Supervised Queries. arXiv preprint arXiv:2409.19548 (2024)

    2024 arXiv

  28. [36]

    Shaowei Yao, Jiwei Tan, Xi Chen, Juhao Zhang, Xiaoyi Zeng, and Keping Yang

  29. [37]

    Andrew Yates, Rodrigo Nogueira, and Jimmy Lin. 2021. Pretrained transformers for text ranking: BERT and beyond. In Proceedings of the 14th ACM International Conference on web search and data mining . 1154–1156

    2021

  30. [38]

    Eric Ye, Xiao Bai, Neil O’Hare, Eliyar Asgarieh, Kapil Thadani, Francisco Perez- Sorrosal, and Sujyothi Adiga. 2022. Multilingual taxonomic web page classifica- tion for contextual targeting at Yahoo. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and ...

    2022

  31. [39]

    Eric Ye, Xiao Bai, Neil O’Hare, Eliyar Asgarieh, Kapil Thadani, Francisco Perez- Sorrosal, and Sujyothi Adiga. 2024. Multilingual Taxonomic Web Page Categoriza- tion Through Ensemble Knowledge Distillation. IEEE Transactions on Knowledge and Data Engineering (2024)

    2024

  32. [40]

    Yukun Zheng, Jiang Bian, Guanghao Meng, Chao Zhang, Honggang Wang, Zhixuan Zhang, Sen Li, Tao Zhuang, Qingwen Liu, and Xiaoyi Zeng. 2022. Multi-Objective Personalized Product Retrieval in Taobao Search. arXiv preprint arXiv:2210.04170 (2022)

    2022 arXiv

  33. [2022]

    In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

    ReprBERT: distilling BERT to an efficient representation-based relevance model for e-commerce. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . 4363–4371

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.