Pith. sign in

REVIEW 3 major objections 4 minor 47 references

TACLR: A Scalable and Efficient Retrieval-based Method for Industrial Product Attribute Value Identification

T0 review · 3 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read This paper claims that product attribute value identification can be reformulated as retrieval over a taxonomy, and that the resulting contrastively trained retriever, TACLR, outperforms fine-tuned LLMs on two benchmarks while running at…

desk verdict Retrieval-based PAVI is a real contribution with good efficiency and a strong Xianyu result, but the OOD claim needs reframing and the WDC comparison is cherry-picked; worth refereeing after fixes. read the letter →

arxiv 2501.03835 v4 pith:KSVSFCYA submitted 2025-01-07 cs.CL cs.AIcs.IR

Yindu Su , Huike Zou , Lin Sun , Ting Zhang , Haiyang Yang , Liyu Chen , David Lo , Qingheng Zhang
show 2 more authors
Shuguang Han Jufeng Chen
This is my paper · ORCID
classification cs.CLcs.AIcs.IR
keywords productattributevalueidentificationretrieval-basedmethodcontrastivelearningtaxonomy-awarenegativesamplingadaptivethresholdnullembeddingout-of-distributionvaluese-commercedata
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

Product attribute value identification (PAVI)—deciding, for example, that a listing that says "iphone12pm256, basically brand new" has Model "iPhone 12 Pro Max", Capacity "256GB", and Brand "Apple" even though "Apple" never appears in the text—is usually done by extraction, classification, or LLM generation, and each paradigm gives up something important: implicit values, out-of-distribution values, normalized output, or speed. This paper tries to establish that all four requirements can be met at once by treating PAVI as information retrieval: the product profile is the query and the attribute taxonomy is the corpus of candidate values. The proposed method, TACLR, trains a shared text encoder with contrastive learning, using hard negatives drawn from the same category-attribute pair, and at inference uses a learned null value per attribute as a dynamic threshold to decide when no value applies. If the paper is right, e-commerce platforms can replace slow LLM-based value generation with a fast vector retrieval step that produces normalized values and tolerates paraphrases and implicit mentions.

What carries the argument

The load-bearing machinery is a contrastively trained dual-encoder retriever. A single text encoder (a pre-trained language model with a linear projection head) embeds the product profile, formatted as "title: ... description: ...", and each candidate value, formatted with a context-rich prompt "A {category} with {attribute} being {value}". Training uses a contrastive loss in which negatives are not random in-batch samples but hard negatives sampled from the same category-attribute value set, forcing the model to distinguish semantically close values like iPhone 12 Pro from iPhone 13 Pro Max. For every category-attribute pair a learnable null value $v_0^a$ is trained alongside the real values; at inference its embedding similarity to the item acts as a dynamic threshold, so "no value for this attribute" is decided by the same retrieval mechanism instead of a hand-tuned cutoff. This design is what lets the paper claim normalized outputs (candidates come from the taxonomy), implicit-value handling (semantic rather than lexical matching), and scalable inference (value embeddings are precomputed and indexed).

What would settle it

Freeze the attribute taxonomy at a fixed date, select product profiles whose correct attribute values are items released after that date (for example, a phone model introduced post-freeze), and run TACLR with the frozen taxonomy. If its top-1 recall for those profiles is near zero while a fine-tuned LLM baseline that reads the same profile can name the new value, the paper's claim that TACLR generalizes to out-of-distribution values is false; the method would only be matching against values that were already known.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a retrieval-based formulation of PAVI is not just a practical speedup but a better model of the task. TACLR encodes the product profile and every candidate value in the attribute taxonomy into a shared embedding space, retrieves the top-1 value per attribute by cosine similarity, and uses the similarity to a learned null value as a per-attribute adaptive threshold. On Xianyu-PAVI, a proprietary large-scale dataset with 8,803 categories and 6.3 million category-attribute-value tuples, TACLR reaches an F1 of 86.2%, above fine-tuned Llama3.1 (84.7%) and Qwen2.5 (81.7%); on WDC-PAVE, excluding measurement attributes that require unit reasoning, it reaches 80.3%, matching or slightly exceeding LLM RAG baselines. It also scores higher on the implicit and unnormalized subset (82.9% versus 79.4% and 78.6%) and runs at 630 samples per second on a single V100, compared with under 100 for the LLM generation baselines.

Load-bearing premise

The method only works for values that already exist in the maintained attribute taxonomy, so genuinely new values that nobody has added to the taxonomy yet are invisible to it.

Editorial extensions

If this is right

  • PAVI becomes a vector-search problem: value embeddings are precomputed offline and indexed, so serving a new item is one encoder pass plus nearest-neighbor search, without autoregressive decoding.
  • Normalized output is free because the system can only return values that exist in the taxonomy as candidates.
  • Implicit and paraphrased values are handled by semantic similarity, which the paper's subset analysis supports with an F1 of 82.9% on unnormalized and implicit pairs.
  • The dynamic null threshold removes per-pair threshold tuning and adapts automatically as the taxonomy grows, since each null embedding is learned per category-attribute pair.
  • If a value is not in the taxonomy, it cannot be retrieved; the paper's own Limitations section states that taxonomy maintenance remains manual, so out-of-distribution generalization is bounded by taxonomy coverage.

Reading between the lines

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

  • The paper does not quantify how quickly retrieval quality decays as the taxonomy ages; a direct test would freeze a taxonomy at a past date, evaluate on products released afterward, and measure F1 loss per month of missing values, which would put a number on the manual maintenance cost the paper concedes.
  • The same dual-encoder-plus-null-threshold recipe transfers to any structured prediction task with a fixed hierarchical label set, such as product categorization or attribute tagging in other domains, where hard negatives can be drawn from sibling labels.
  • The learned null embeddings could be probed directly: if they cluster by attribute rather than by category, then the "dynamic threshold" is really encoding attribute-level unrelatedness rather than a category-aware decision boundary, which would refine how the method is described and extended.
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 / 4 minor

Summary. The paper proposes TACLR, a retrieval-based method for Product Attribute Value Identification (PAVI). It encodes product profiles and candidate values with a shared text encoder, trains a contrastive objective with taxonomy-aware hard negative sampling, and learns a null-value embedding per category-attribute pair that serves as a dynamic threshold at inference. The method is evaluated on a large proprietary Xianyu dataset and on the public WDC-PAVE dataset against classification, fine-tuned LLM, and RAG baselines, reporting F1 86.2% on Xianyu and F1 Excl. 80.3% on WDC-PAVE, a throughput of 630 samples/second, and a production deployment at Xianyu. The code is released.

Significance. If the empirical claims hold, TACLR is a valuable industrial contribution. The retrieval formulation is a sensible way to produce normalized outputs at scale, the taxonomy-aware negative sampling strategy is a reasonable adaptation of contrastive learning to this task, and the learned null-value threshold is an interesting alternative to fixed per-pair cutoffs. Releasing the code and reporting deployment details are strengths that support reproducibility and practical adoption. However, the advertised claims of out-of-distribution generalization and superior WDC-PAVE performance are currently supported only in a narrower sense than the abstract suggests: retrieved values must already exist in the maintained taxonomy, and on the full WDC test set TACLR's F1 is below several LLM-based baselines.

major comments (3)
  1. [§3.1, §3.2, §5.3 (Fig. 5c)] The 'out-of-distribution' claim is narrower than advertised and rests on an underspecified evaluation. Section 3.1 defines PAVI as selecting values from a predefined taxonomy V_a, and the footnote there explicitly states that the method leverages an existing taxonomy rather than constructing or updating it; Section 3.2 precomputes embeddings of all taxonomy values. Therefore the output space at inference is exactly V_a plus the null value. The cross-value split in Figure 5c is the only direct evidence for OOD behavior, but the paper does not describe how held-out values were selected or whether those values were already in the taxonomy with precomputed embeddings during evaluation. If they were, the result demonstrates generalization to unseen training labels, not to taxonomy entries absent at test time. Please either revise the abstract and contributions to say 'values unseen in training but present in the taxonomy' or add an evaluation that simulates taxonomy updates by inserting genuinely new values at inference. Section 7's concession that taxonomy maintenance requires ongoing manual updates further supports this narrower reading.
  2. [§5.1, Table 4] The WDC-PAVE result is reported in a way that obscures the full comparison. TACLR's full WDC F1 is 72.6, which is below Llama3.1-RAG (77.2), Llama3.1-few-shot (75.0), and Qwen2.5-RAG (74.2); TACLR is only above these baselines when measurement attributes are excluded (F1 Excl. 80.3). The main-text statement that TACLR 'achieves the highest F1 Excl. score' should be accompanied by the full F1 and a clear acknowledgment that on the complete test set TACLR is not superior. Since Section 7 identifies measurement attributes as a limitation of retrieval-based methods, excluding them from the headline metric is legitimate only if the limitation is prominently disclosed and the full result is equally visible.
  3. [Table 4, Appendix A] The central comparison on Xianyu-PAVI is reported from a single run without variance or significance information. The 1.5-point gap between TACLR (86.2 F1) and fine-tuned Llama3.1 (84.7 F1) could plausibly lie within run-to-run noise, especially since hyperparameters and checkpoints are selected by validation F1 and no seeds are reported. Please provide multiple seeds or a small number of repeats for the main comparison, or otherwise justify the stability of the reported superiority.
minor comments (4)
  1. [Table 4] Several cells in Table 4 contain merged or missing numbers (for example, '47.678.2' and '77.280.1' in the Llama3.1-RAG row); please fix the table so that Precision, Recall, and F1 are clearly separated.
  2. [§3.4] The phrase 'dynamic thresholds' is somewhat overstated, since the threshold is a null-value embedding learned during training rather than a quantity that adapts at test time. Consider describing it as a learned per-pair threshold to avoid the implication of online adaptation.
  3. [Figure 5c] The caption says 'data domains,' but the figure actually compares in-domain, cross-category, and cross-value splits; please align the caption with the figure content.
  4. [Throughout] The notation for the task name alternates between 'PAVI' and 'PA VI' (also 'PAVE' versus 'PA VE'); please use a single consistent form throughout the text and figures.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: TACLR is an empirical retrieval pipeline whose predictions are evaluated on held-out labels, and its OOD claim is explicitly scoped to the provided taxonomy.

full rationale

TACLR does not present a derivation of a result from its own assumptions; it is a supervised retrieval system. The item and value encoders, contrastive loss, and null-value-based dynamic threshold are trained on labeled product-attribute pairs and evaluated on held-out test data, including the public WDC-PAVE dataset. The headline F1 scores are computed against external ground-truth labels and compared with classification, generation, and RAG baselines; no reported score is the fitted value of a parameter being presented as a prediction. The dynamic threshold in Section 3.4 is a learned null-value embedding, but the actual prediction is the argmax over the candidate value set plus the null value, and the threshold's effect is measured on validation and test data rather than being tuned to reproduce the reported F1. The paper's claim about handling out-of-distribution values is explicitly bounded in Section 3.1, which states that the approach leverages an existing attribute taxonomy rather than constructing or updating it, and Section 7 concedes that taxonomy maintenance requires ongoing manual updates. This makes the OOD claim narrower than the abstract might suggest, but it is a scope limitation, not a circular reduction. The self-citations that appear (e.g., Chen et al., 2024, in the introduction) are contextual references to related product-listing work and are not load-bearing for the proposed method's derivation or evaluation. Overall, the evaluation is self-contained against external benchmarks, and no equation or fitted quantity reduces by construction to the claim being tested.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The behavior of TACLR rests on a maintained attribute taxonomy, a top-1 exclusive-value assumption, and a learned null embedding per category-attribute pair that acts as the decision threshold. These are inputs or fitted components, not derived results. No new physical or conceptual entities are introduced; the null value is a model parameter.

free parameters (3)
  • Temperature tau = 0.05
    Fixed based on prior work and validation tuning (Appendix A). It scales the contrastive loss and affects the distribution of similarities.
  • Number of sampled negatives K = 128
    Hyperparameter selected by validation; F1 improves with sample size (Figure 4), but 128 is the maximum tested, not a theoretically determined value.
  • Null value embeddings per category-attribute pair = Learned vectors, one per CA pair (26,645 on Xianyu)
    These embeddings define the dynamic threshold and are directly fitted to the training data. They are the mechanism that turns a learned similarity score into a decision boundary, so the 'adaptive threshold' is a fitted model component.
assumptions (4)
  • domain assumption A predefined attribute taxonomy exists and covers the relevant categories, attributes, and values.
    The task definition in Section 3.1 assumes the taxonomy is given, and the Limitations section (Section 7) states that the method requires ongoing manual taxonomy updates.
  • domain assumption Most category-attribute pairs have exclusive values, so predicting top-1 is sufficient.
    Section 3.4 states 'most category-attribute pairs have exclusive values' and explicitly focuses on top-1 prediction, which leaves multi-valued attributes out of scope.
  • domain assumption A shared text encoder can embed both product profiles and candidate values into a space where cosine similarity identifies implicit and unnormalized values.
    This is the core modeling hypothesis of the paper. It is evaluated empirically, but not proven, and the paper does not analyze failure modes of the embedding space.
  • domain assumption The manual annotations in Xianyu-PAVI are reliable enough to serve as ground truth.
    Section 4.1 describes a multi-stage annotation process but reports no inter-annotator agreement, so label noise is not quantified.

how reviews work

0 comments
Cite this review

Pith. "Pith review of TACLR: A Scalable and Efficient Retrieval-based Method for Industrial Product Attribute Value Identification." pith.science (2026). https://pith.science/paper/KSVSFCYA

@misc{pith2026250103835,
  author       = {Pith},
  title        = {Pith review of: TACLR: A Scalable and Efficient Retrieval-based Method for Industrial Product Attribute Value Identification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KSVSFCYA}},
  note         = {Machine review of arXiv:2501.03835}
}
read the original abstract

Product Attribute Value Identification (PAVI) involves identifying attribute values from product profiles, a key task for improving product search, recommendation, and business analytics on e-commerce platforms. However, existing PAVI methods face critical challenges, such as inferring implicit values, handling out-of-distribution (OOD) values, and producing normalized outputs. To address these limitations, we introduce Taxonomy-Aware Contrastive Learning Retrieval (TACLR), the first retrieval-based method for PAVI. TACLR formulates PAVI as an information retrieval task by encoding product profiles and candidate values into embeddings and retrieving values based on their similarity. It leverages contrastive training with taxonomy-aware hard negative sampling and employs adaptive inference with dynamic thresholds. TACLR offers three key advantages: (1) it effectively handles implicit and OOD values while producing normalized outputs; (2) it scales to thousands of categories, tens of thousands of attributes, and millions of values; and (3) it supports efficient inference for high-load industrial deployment. Extensive experiments on proprietary and public datasets validate the effectiveness and efficiency of TACLR. Further, it has been successfully deployed on the real-world e-commerce platform Xianyu, processing millions of product listings daily with frequently updated, large-scale attribute taxonomies. We release the code to facilitate reproducibility and future research at https://github.com/SuYindu/TACLR.

Figures

Figures reproduced from arXiv: 2501.03835 by the authors.

Figure 1
Figure 1. Illustration of the PAVE and PAVI tasks. Un [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. An illustration of a portion of the attribute [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Overview of the training and inference process of TACLR, our retrieval-based method for the PAVI task. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Comparison of negative sampling strategies [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Performance analysis across inference thresholds, prompt templates, and data domains. [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 19 canonical work pages

  1. [1]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRING...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Ansel Blume, Nasser Zalmout, Heng Ji, and Xian Li. 2023. https://doi.org/10.18653/v1/2023.emnlp-industry.55 Generative models for product attribute extraction . In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 575--585, Singapore. Association for Computational Linguistics

    doi:10.18653/v1/2023.emnlp-industry.55 2023

  4. [4]

    Alexander Brinkmann, Nick Baumann, and Christian Bizer. 2024. https://doi.org/10.1007/978-3-031-70626-4\_15 Using llms for the extraction and normalization of product attribute values . In European Conference on Advances in Databases and Information Systems, pages 217--230. Springer

    work page doi:10.1007/978-3-031-70626-4 2024

  5. [5]

    Kang Chen, Qing Heng Zhang, Chengbao Lian, Yixin Ji, Xuwei Liu, Shuguang Han, Guoqiang Wu, Fei Huang, and Jufeng Chen. 2024. https://doi.org/10.18653/v1/2024.emnlp-industry.52 IPL : Leveraging multimodal large language models for intelligent product listing . In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Indust...

    work page doi:10.18653/v1/2024.emnlp-industry.52 2024

  6. [6]

    Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. 2020. https://proceedings.mlr.press/v119/chen20j.html A simple framework for contrastive learning of visual representations . In Proceedings of the 37th International Conference on Machine Learning, volume 119 of Proceedings of Machine Learning Research, pages 1597--1607. PMLR

    work page 2020

  7. [7]

    Wei-Te Chen, Keiji Shinzato, Naoki Yoshinaga, and Yandi Xia. 2023. https://doi.org/10.18653/v1/2023.emnlp-industry.16 Does named entity recognition truly not scale up to real-world product attribute extraction? In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 152--159, Singapore. Association ...

    work page doi:10.18653/v1/2023.emnlp-industry.16 2023

  8. [8]

    Wei-Te Chen, Yandi Xia, and Keiji Shinzato. 2022. https://doi.org/10.18653/v1/2022.ecnlp-1.16 Extreme multi-label classification with label masking for product attribute value extraction . In Proceedings of the Fifth Workshop on e-Commerce and NLP (ECNLP 5), pages 134--140, Dublin, Ireland. Association for Computational Linguistics

    work page doi:10.18653/v1/2022.ecnlp-1.16 2022

Show all 47 references
  1. [9]

    Yiming Cui, Wanxiang Che, Ting Liu, Bing Qin, Shijin Wang, and Guoping Hu. 2020. https://doi.org/10.18653/v1/2020.findings-emnlp.58 Revisiting pre-trained models for C hinese natural language processing . In Findings of the Association for Computational Linguistics: EMNLP 2020...

    2020 doi

  2. [10]

    Yifan Ding, Yan Liang, Nasser Zalmout, Xian Li, Christan Grant, and Tim Weninger. 2022. https://doi.org/10.18653/v1/2022.emnlp-industry.9 Ask-and-verify: Span candidate generation and verification for attribute value extraction . In Proceedings of the 2022 Conference on Empiri...

    2022 doi

  3. [11]

    Shen Gao, Zhaochun Ren, Yihong Zhao, Dongyan Zhao, Dawei Yin, and Rui Yan. 2019. https://doi.org/10.1145/3289600.3290992 Product-aware answer generation in e-commerce question-answering . In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining,...

    2019

  4. [12]

    Rayid Ghani, Katharina Probst, Yan Liu, Marko Krema, and Andrew Fano. 2006. https://doi.org/10.1145/1147234.1147241 Text mining for product attribute extraction . SIGKDD Explor. Newsl., 8(1):41–48

    2006

  5. [13]

    Kaiming He, Haoqi Fan, Yuxin Wu, Saining Xie, and Ross Girshick. 2020. https://doi.org/10.1109/CVPR42600.2020.00975 Momentum contrast for unsupervised visual representation learning . In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 9726--9735

    2020

  6. [14]

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. https://openreview.net/forum?id=nZeVKeeFYf9 Lo RA : Low-rank adaptation of large language models . In International Conference on Learning Representations

    2022

  7. [15]

    Junshi Huang, Si Liu, Junliang Xing, Tao Mei, and Shuicheng Yan. 2014. https://doi.org/10.1145/2632165 Circle & search: Attribute-aware shoe retrieval . ACM Transactions on Multimedia Computing, Communications, and Applications, 11(1)

    2014 doi

  8. [16]

    Lei Huang, Weijiang Yu, Weitao Ma, Weihong Zhong, Zhangyin Feng, Haotian Wang, Qianglong Chen, Weihua Peng, Xiaocheng Feng, Bing Qin, and Ting Liu. 2024. https://doi.org/10.1145/3703155 A survey on hallucination in large language models: Principles, taxonomy, challenges, and o...

    2024 doi

  9. [17]

    Giannis Karamanolakis, Jun Ma, and Xin Luna Dong. 2020. https://doi.org/10.18653/v1/2020.acl-main.751 TX tract: Taxonomy-aware knowledge extraction for thousands of product categories . In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics,...

    2020 doi

  10. [18]

    Anant Khandelwal, Happy Mittal, Shreyas Kulkarni, and Deepak Gupta. 2023. https://doi.org/10.18653/v1/2023.acl-industry.29 Large scale generative multimodal attribute extraction for E -commerce attributes . In Proceedings of the 61st Annual Meeting of the Association for Compu...

    2023 doi

  11. [19]

    Ashish Kulkarni, Kartik Mehta, Shweta Garg, Vidit Bansal, Nikhil Rasiwasia, and Srinivasan Sengamedu. 2019. https://doi.org/10.1145/3308560.3316597 Productqna: Answering user questions on e-commerce product pages . In Companion Proceedings of The 2019 World Wide Web Conference...

    2019

  12. [20]

    Yanzeng Li, Bingcong Xue, Ruoyu Zhang, and Lei Zou. 2023. https://doi.org/10.18653/v1/2023.acl-long.119 A t TG en: Attribute tree generation for real-world attribute joint extraction . In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (...

    2023 doi

  13. [21]

    Rongmei Lin, Xiang He, Jie Feng, Nasser Zalmout, Yan Liang, Li Xiong, and Xin Luna Dong. 2021. https://doi.org/10.1145/3447548.3467164 Pam: Understanding product images in cross product category attribute extraction . In Proceedings of the 27th ACM SIGKDD Conference on Knowled...

    2021

  14. [22]

    Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. https://api.semanticscholar.org/CorpusID:198953378 Roberta: A robustly optimized bert pretraining approach . ArXiv, abs/1907.11692

    2019 arXiv

  15. [23]

    Meta Llama Team. 2024. https://arxiv.org/abs/2407.21783 The llama 3 herd of models . Preprint, arXiv:2407.21783

    2024 arXiv

  16. [24]

    Alessandro Magnani, Feng Liu, Min Xie, and Somnath Banerjee. 2019. https://doi.org/10.1145/3308560.3316603 Neural product retrieval at walmart.com . In Companion Proceedings of The 2019 World Wide Web Conference, WWW '19, page 367–372, New York, NY, USA. Association for Comput...

    2019

  17. [25]

    Nikolakopoulos, Swati Kaul, Siva Karthik Gade, Bella Dubrov, Umit Batur, and Suleiman Ali Khan

    Athanasios N. Nikolakopoulos, Swati Kaul, Siva Karthik Gade, Bella Dubrov, Umit Batur, and Suleiman Ali Khan. 2023. https://arxiv.org/abs/2309.05920 Sage: Structured attribute value generation for billion-scale product catalogs . Preprint, arXiv:2309.05920

    2023 arXiv

  18. [26]

    Duangmanee Putthividhya and Junling Hu. 2011. https://aclanthology.org/D11-1144 Bootstrapped named entity recognition for product attribute extraction . In Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, pages 1557--1567, Edinburgh, Scot...

    2011

  19. [27]

    Alibaba Qwen Team. 2024. https://arxiv.org/abs/2407.10671 Qwen2 technical report . Preprint, arXiv:2407.10671

    2024 arXiv

  20. [28]

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. 2021. https://proceedings.mlr.press/v139/radford21a.html Learning transferable visual model...

    2021

  21. [29]

    Kalyani Roy, Pawan Goyal, and Manish Pandey. 2021. https://doi.org/10.18653/v1/2021.ecnlp-1.2 Attribute value generation from product title using language models . In Proceedings of the 4th Workshop on e-Commerce and NLP, pages 13--17, Online. Association for Computational Linguistics

    2021 doi

  22. [30]

    Kassem Sabeh, Mouna Kacimi, Johann Gamper, Robert Litschko, and Barbara Plank. 2024 a . https://arxiv.org/abs/2409.12695 Exploring large language models for product attribute value identification . Preprint, arXiv:2409.12695

    2024 arXiv

  23. [31]

    Kassem Sabeh, Robert Litschko, Mouna Kacimi, Barbara Plank, and Johann Gamper. 2024 b . https://arxiv.org/abs/2407.01137 An empirical comparison of generative approaches for product attribute-value identification . Preprint, arXiv:2407.01137

    2024 arXiv

  24. [32]

    Keiji Shinzato, Naoki Yoshinaga, Yandi Xia, and Wei-Te Chen. 2022. https://doi.org/10.18653/v1/2022.acl-short.25 Simple and effective knowledge-driven query expansion for QA -based product attribute extraction . In Proceedings of the 60th Annual Meeting of the Association for ...

    2022 doi

  25. [33]

    Keiji Shinzato, Naoki Yoshinaga, Yandi Xia, and Wei-Te Chen. 2023. https://doi.org/10.18653/v1/2023.findings-acl.413 A unified generative approach to product attribute-value identification . In Findings of the Association for Computational Linguistics: ACL 2023, pages 6599--66...

    2023 doi

  26. [34]

    Changfeng Sun, Han Liu, Meng Liu, Zhaochun Ren, Tian Gan, and Liqiang Nie. 2020. https://doi.org/10.1145/3336191.3371805 Lara: Attribute-to-feature adversarial learning for new-item recommendation . In Proceedings of the 13th International Conference on Web Search and Data Min...

    2020

  27. [36]

    Sivakumar, Bin Shu, Zac Yu, and Jon Elsas

    Qifan Wang, Li Yang, Bhargav Kanagal, Sumit Sanghai, D. Sivakumar, Bin Shu, Zac Yu, and Jon Elsas. 2020. https://doi.org/10.1145/3394486.3403047 Learning to extract attribute value from product via question answering: A multi-task approach . In Proceedings of the 26th ACM SIGK...

    2020

  28. [37]

    Qifan Wang, Li Yang, Jingang Wang, Jitin Krishnan, Bo Dai, Sinong Wang, Zenglin Xu, Madian Khabsa, and Hao Ma. 2022. https://doi.org/10.18653/v1/2022.findings-emnlp.20 SMARTAVE : Structured multimodal transformer for product attribute value extraction . In Findings of the Asso...

    2022 doi

  29. [38]

    Shitao Xiao, Zheng Liu, Peitian Zhang, and Niklas Muennighoff. 2023. https://arxiv.org/abs/2309.07597 C-pack: Packaged resources to advance general chinese embedding . Preprint, arXiv:2309.07597

    2023 arXiv

  30. [39]

    Huimin Xu, Wenting Wang, Xin Mao, Xinyu Jiang, and Man Lan. 2019. https://doi.org/10.18653/v1/P19-1514 Scaling up open tagging from tens to thousands: Comprehension empowered attribute value extraction from product title . In Proceedings of the 57th Annual Meeting of the Assoc...

    2019 doi

  31. [40]

    Liyan Xu, Chenwei Zhang, Xian Li, Jingbo Shang, and Jinho D. Choi. 2023. https://doi.org/10.18653/v1/2023.acl-long.683 Towards open-world product attribute mining: A lightly-supervised approach . In Proceedings of the 61st Annual Meeting of the Association for Computational Li...

    2023 doi

  32. [41]

    Jun Yan, Nasser Zalmout, Yan Liang, Christan Grant, Xiang Ren, and Xin Luna Dong. 2021. https://doi.org/10.18653/v1/2021.acl-long.362 A da T ag: Multi-attribute value extraction from product profiles with adaptive decoding . In Proceedings of the 59th Annual Meeting of the Ass...

    2021 doi

  33. [42]

    Li Yang, Qifan Wang, Jingang Wang, Xiaojun Quan, Fuli Feng, Yu Chen, Madian Khabsa, Sinong Wang, Zenglin Xu, and Dongfang Liu. 2023. https://doi.org/10.18653/v1/2023.findings-acl.633 M ix PAVE : Mix-prompt tuning for few-shot product attribute value extraction . In Findings of...

    2023 doi

  34. [43]

    Li Yang, Qifan Wang, Zac Yu, Anand Kulkarni, Sumit Sanghai, Bin Shu, Jon Elsas, and Bhargav Kanagal. 2022. https://doi.org/10.1145/3488560.3498377 Mave: A product dataset for multi-source attribute value extraction . In Proceedings of the Fifteenth ACM International Conference...

    2022

  35. [44]

    Danqing Zhang, Zheng Li, Tianyu Cao, Chen Luo, Tony Wu, Hanqing Lu, Yiwei Song, Bing Yin, Tuo Zhao, and Qiang Yang. 2021. https://doi.org/10.1145/3459637.3481946 Queaco: Borrowing treasures from weakly-labeled behavior data for query attribute value extraction . In Proceedings...

    2021

  36. [45]

    Xinyang Zhang, Chenwei Zhang, Xian Li, Xin Luna Dong, Jingbo Shang, Christos Faloutsos, and Jiawei Han. 2022. https://doi.org/10.1145/3485447.3512035 Oa-mine: Open-world attribute mining for e-commerce products with weak supervision . In Proceedings of the ACM Web Conference 2...

    2022

  37. [46]

    Guineng Zheng, Subhabrata Mukherjee, Xin Luna Dong, and Feifei Li. 2018. https://doi.org/10.1145/3219819.3219839 Opentag: Open attribute value extraction from product profiles . In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining...

    2018

  38. [47]

    Tiangang Zhu, Yue Wang, Haoran Li, Youzheng Wu, Xiaodong He, and Bowen Zhou. 2020. https://doi.org/10.18653/v1/2020.emnlp-main.166 Multimodal joint attribute prediction and value extraction for E -commerce product . In Proceedings of the 2020 Conference on Empirical Methods in...

    2020 doi

  39. [48]

    Henry Zou, Gavin Yu, Ziwei Fan, Dan Bu, Han Liu, Peng Dai, Dongmei Jia, and Cornelia Caragea. 2024. https://doi.org/10.18653/v1/2024.naacl-industry.40 EIVEN : Efficient implicit attribute value extraction using multimodal LLM . In Proceedings of the 2024 Conference of the Nort...

    2024 doi

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.