Pith. sign in

REVIEW 3 major objections 6 minor 58 references

Generative Retrieval for Book search

T0 review · 3 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Pith's one-line read A new generative retrieval framework for book search, built on outline-based encoding and LLM-augmented training data, reports gains over the strongest prior generative baseline on both industry and public book datasets.

desk verdict Solid industrial GR-for-books paper, but the headline BBS gain is inflated by a same-distribution synthetic test set; the public real-query result is the more honest number. read the letter →

arxiv 2501.11034 v1 pith:RMH7XVKM submitted 2025-01-19 cs.IR

classification cs.IR
keywords generativeretrievalbooksearchoutline-orientedencodingpseudo-queryaugmentationidentifierslong-textmodelingretentiveattentionpositional
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

Book search differs from web search because a book's relevant content is spread across metadata, a hierarchical outline, and a very long main text. This paper argues that generative retrieval, a paradigm that compresses the whole corpus into one model that directly emits document identifiers for a query, can be made to work for books if two problems are solved: the model needs enough training signal to cover a book's many facets, and it needs an encoding that respects the outline's chapter-section structure. The proposed framework, GBS, solves the first with identifier augmentation and pseudo-queries generated by large language models, and the second with outline-oriented positional encodings and a retentive attention memory. On a proprietary 40K-book dataset it reports a 9.8% gain in MRR@20 over the strongest prior generative baseline, and on a public dataset of real book queries it reports a 6% gain in Hits@10. The sympathetic reading is that hierarchy-aware encoding and augmented training pairs, not just scale, are what let generative retrieval handle books.

What carries the argument

The load-bearing machinery is the pair of augmentation schemes plus the outline-oriented encoder. Coverage-promoting identifier augmentation constructs book-, chapter-, and section-level identifiers and ties each to the content form that matches its granularity, so the model learns the mapping from whole-book and partial views to a shared identifier. Diversity-enhanced query augmentation uses two LLM prompts to generate single-chapter answerable and multiple-chapter answerable pseudo-queries, giving the retrieval task a spread of difficulty. The encoder applies bi-level positional encoding, with absolute positions inside sections and rotary relative positions between sections, and outline-oriented retentive attention, which augments standard multi-head attention with a memory state updated as a linear-attention recurrence. These mechanisms together are what the paper credits for the reported gains over RIPOR.

What would settle it

Build a held-out test set from real user queries to a live book search service, with no LLM-generated queries, and compare GBS against RIPOR on MRR@20 and Hits@10; if the gap shrinks below the reported 9.8% and 6%, the advantage is driven by the pseudo-query distribution.

Watch

Extended reading notes

Core claim

The paper's central discovery is that applying generative retrieval to books fails when the book is treated as flat segments and succeeds when each book is represented by multiple content views, multiple identifier levels, and an outline-aware encoder. GBS builds hierarchical identifiers at book, chapter, and section levels from titles, authors, publishers, and cluster-derived semantic numbers, then trains the indexing task by pairing keywords, summaries, section text, chapter text, and whole text with the appropriate identifier level. For the retrieval task it pairs LLM-generated pseudo-queries, of two difficulty types, with book-level identifiers. The encoder combines section-local positional encodings with chapter-level rotary relative encodings, and adds a retentive memory that carries key-value states across long inputs in a linear-attention recurrent update. The reported numbers are MRR@20 of 46.9 versus 42.7 for the RIPOR baseline on the 40K-book industry dataset, and Hits@10 of 70.7 versus 66.7 on WhatsThatBook, which the paper takes as evidence that hierarchical structure is the enabling ingredient for generative book search.

Load-bearing premise

The main benchmark's test queries are generated by the same LLM prompts used to create the training queries, so the reported advantage over RIPOR presupposes that such pseudo-queries stand in for what real book-search users actually type.

Editorial extensions

If this is right

  • Each component contributes: removing query augmentation drops Hits@10 from 56.7 to 50.6 on the 40K-book dataset, removing identifier augmentation drops it to 45.3, and removing bi-level positional encoding drops it to 52.8.
  • Retentive attention provides a smaller but consistent lift, and its value increases with input length up to 128K tokens, after which performance slightly declines.
  • Parallel decoding with book- and chapter-level prefix trees works slightly better than serial decoding, and both variants outperform every non-generative and generative baseline tested.
  • If the reported results hold, generative retrieval is a viable retrieval paradigm for book-scale documents, and the same outline-aware recipe can be tested on other long hierarchical texts such as legal or technical corpora.

Reading between the lines

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

  • Because the main test set is generated by the same LLM pseudo-query procedure used to create training queries, the 9.8% gain measures performance on the distribution the model was trained for; the WhatsThatBook result, built from real user queries, is the more informative transfer test.
  • The outline-based encoder should transfer to other structured long documents, but the gain would depend on the availability of an explicit hierarchy comparable to a book outline.
  • A natural stress test is to shift the pseudo-query generator, prompt, or language at test time and measure how much of the advantage survives; the paper does not report this.
  • The stated training-cost bottleneck suggests an obvious next step: use the outline to allocate augmentation selectively to chapters or sections where retrieval errors are concentrated, rather than generating pseudo-queries uniformly.
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 proposes GBS, a generative retrieval (GR) framework for book search. GBS combines two main ideas: (i) data augmentation, which includes coverage-promoting book identifier augmentation (multiple content forms mapped to hierarchical book/chapter/section identifiers) and diversity-enhanced query augmentation (LLM-generated pseudo-queries of single-chapter and multiple-chapter types); and (ii) outline-oriented book encoding, which applies bi-level positional encoding and a retentive attention memory to handle long book texts. Training uses multi-task MLE over indexing and retrieval pairs (Eq. 4-6), and inference uses constrained decoding with book- and chapter-level prefix trees and parallel or serial aggregation (Section 2.6). Experiments on a proprietary Baidu BBS dataset (10K/20K/40K books) and the public WhatsThatBook dataset report improvements over the strongest GR baseline RIPOR, e.g., 9.8% MRR@20 on BBS 40K and 6% Hits@10 on WhatsThatBook (Table 2).

Significance. If the reported results are valid, GBS would be a meaningful advance in applying GR to long structured documents and a practical contribution to book search. The paper is clearly written, the method is well motivated, and the ablations (Tables 3 and 4) show that each component contributes. A notable strength is the evaluation on WhatsThatBook, which uses real user queries and provides external grounding beyond the proprietary dataset. However, the headline BBS result is undermined by the evaluation protocol: the BBS test queries are generated with the same LLM prompts and from the same book content used for training, so the 9.8% MRR@20 gain over RIPOR may reflect memorization of the training distribution rather than retrieval generalization. The public WhatsThatBook result shows a smaller gain (6% Hits@10) and is more credible as evidence of real-world effectiveness.

major comments (3)
  1. [Section 3 (Datasets) and Section 2.3.2] The BBS test set is constructed with the same procedure as the training pseudo-queries: Section 3 states that pseudo-queries are generated 'for each book for training and evaluation using the method described in Section 2.3,' i.e., from the same chapter texts and with the same LLM prompts. Because the indexing task (Eq. 4) already trains the model to map whole text, chapter text, section text, keywords, and summaries to identifiers, a test pseudo-query that paraphrases or summarizes a chapter can be answered by recognizing content the model has seen in training and tied to a book-id. This creates a circularity that likely inflates the reported 9.8% MRR@20 improvement over RIPOR on BBS 40K. The WhatsThatBook experiment uses real user queries and shows a smaller 6% Hits@10 gain, consistent with this concern. The manuscript should either re-evaluate BBS with held-out books/queries or explicitly present BBS as a synthetic stress test and base the central claim on the public dataset.
  2. [Section 2.3.1 and Table 5 (case study)] The book-id format is title#author#publisher, and the single-chapter query prompt asks the LLM to generate queries 'focus on key themes, events, characters, and any specific details provided in the chapter.' The case study in Table 5 includes the query 'Who is the author of the book “The Heart of a Boy”?', which already contains the exact title and author, making the correct book-id trivially recoverable from the query itself without any book understanding. This suggests the BBS test set contains queries that are easy for GBS's identifier design and may contribute to the reported performance gap. The authors should report the distribution of query difficulty (e.g., how many test queries contain the title or author) and ideally filter or separately analyze such cases.
  3. [Section 3, Table 1] The BBS dataset is proprietary and not released, so the headline results cannot be independently reproduced. Given that the test queries are synthetically generated and the dataset is unavailable, the central quantitative claim rests on a black-box setting. At minimum, the authors should release the exact prompt templates, the generated test queries, and the code for the full data-generation and evaluation pipeline, so that reviewers and the community can assess and reproduce the BBS results.
minor comments (6)
  1. [Section 3.1] There is a typo: 'During inference We employ the Adam optimizer' should be 'During inference, we employ the Adam optimizer...' (missing comma and capital letter).
  2. [Section 4.5, Table 5] For the multiple-chapter answerable query 'Introducing Enrico', GBS_P ranks the correct book second (An Introduction to the Basics of Reliability and Risk Analysis), yet the text says 'GBS_P also performs well' without noting that it did not achieve the top rank; this should be stated precisely.
  3. [Figure 4 caption] The caption reads 'Hit@10' but the metric elsewhere is 'Hits@10'; please make the terminology consistent.
  4. [Introduction, paragraph 5] The text refers to 'our proposed method GBR' but the method is named GBS; this is likely a typo and should be corrected.
  5. [Table 2] The dagger symbol (†) indicates p≤0.05 for GBS_P versus RIPOR, but the paper does not specify which statistical test was used, how many runs or folds, or whether the comparison is paired. This should be documented in the experimental settings.
  6. [Section 2.6] The parallel decoding aggregation multiplies book- and chapter-level scores while serial decoding uses a weighted sum; the text cites [17] for the parallel form but gives no justification for the differing aggregation strategies. Please clarify the rationale.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the BBS result rests on a disclosed synthetic-query evaluation, but no prediction is equivalent to its inputs by construction, and the WhatsThatBook real-query benchmark independently supports the method.

full rationale

The claimed derivation chain—indexing with coverage-promoting identifier augmentation, retrieval with diversity-enhanced pseudo-query augmentation, and outline-oriented encoding—is not defined in terms of the reported results. The BBS test queries are generated by the same LLM prompts used to create training queries (Section 3: 'We construct pseudo-queries for each book for training and evaluation using the method described in Section 2.3'), but the specific test query strings are not the training instances, and all baselines are evaluated on the same BBS test set. The relevance labels for pseudo-queries are the source books by construction, which limits external validity for real user queries, but it does not make the model's output equal to its input or reduce the comparison to a fitted parameter. The WhatsThatBook experiment uses real tip-of-the-tongue user queries and reports a 6% Hits@10 improvement over RIPOR, providing independent external support. Section 6's limitation discussion does not flag the synthetic BBS test set, which is an omission in the paper's self-assessment, but it is not a circular step. No load-bearing self-citation, imported uniqueness theorem, or ansatz-by-citation was found; citations to the authors' prior GR work are contextual and not used to justify the central claim.

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

The GBS result depends on several borrowed components and assumptions: pseudo-query quality from LLMs, the availability and reliability of book outlines, the usefulness of k-means-derived semantic identifiers, and prior linear-attention mechanisms. These are not newly introduced entities, but they are premises the empirical claim relies on.

free parameters (4)
  • Number of clusters K for semantic numbers = 10
    Used for chapter- and section-level semantic structured numbers; adopted from DSI [36] without task-specific tuning.
  • Number of pseudo-queries X per type per book = 5
    Main results use X=5; Figure 4 shows performance saturates around X=5, indicating this value was tuned on the BBS evaluation set.
  • Serial decoding aggregation weights beta and gamma = beta=1, gamma=0.5
    Set by hand in Section 3.1; no sensitivity analysis is reported.
  • Input truncation length for whole-text pairs = 128K tokens (first 100 chapters)
    Section 3.1 truncates book input to 128K tokens; Figure 3 shows performance depends on this length.
assumptions (5)
  • domain assumption Book outlines are available for all books and accurately reflect chapter-section structure
    Section 2.3.1 and Section 2.4 construct identifiers and positional encodings from outlines; if outlines are missing or noisy, the method degrades.
  • domain assumption LLM-generated pseudo-queries are representative of real user queries
    Section 3 uses pseudo-queries for both training and BBS evaluation; the headline result is valid only if these queries approximate real search behavior.
  • domain assumption Hierarchical k-means clustering of chapter/section text yields meaningful semantic identifiers
    Section 2.3.1 borrows the DSI identifier scheme [36]; the model must learn content-to-identifier mappings that are discriminative.
  • standard math The retentive memory update equations from [15] are numerically stable and transfer to this setting
    Section 2.4.2 uses Eqs. (1)-(3) from prior work without new analysis.
  • standard math Transformer encoder-decoder with MLE training is an appropriate model for generative retrieval
    Follows DSI [36] and related GR work; no new theoretical foundation is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Generative Retrieval for Book search." pith.science (2026). https://pith.science/paper/RMH7XVKM

@misc{pith2026250111034,
  author       = {Pith},
  title        = {Pith review of: Generative Retrieval for Book search},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RMH7XVKM}},
  note         = {Machine review of arXiv:2501.11034}
}
read the original abstract

In book search, relevant book information should be returned in response to a query. Books contain complex, multi-faceted information such as metadata, outlines, and main text, where the outline provides hierarchical information between chapters and sections. Generative retrieval (GR) is a new retrieval paradigm that consolidates corpus information into a single model to generate identifiers of documents that are relevant to a given query. How can GR be applied to book search? Directly applying GR to book search is a challenge due to the unique characteristics of book search: The model needs to retain the complex, multi-faceted information of the book, which increases the demand for labeled data. Splitting book information and treating it as a collection of separate segments for learning might result in a loss of hierarchical information. We propose an effective Generative retrieval framework for Book Search (GBS) that features two main components: data augmentation and outline-oriented book encoding. For data augmentation, GBS constructs multiple query-book pairs for training; it constructs multiple book identifiers based on the outline, various forms of book contents, and simulates real book retrieval scenarios with varied pseudo-queries. This includes coverage-promoting book identifier augmentation, allowing the model to learn to index effectively, and diversity-enhanced query augmentation, allowing the model to learn to retrieve effectively. Outline-oriented book encoding improves length extrapolation through bi-level positional encoding and retentive attention mechanisms to maintain context over long sequences. Experiments on a proprietary Baidu dataset demonstrate that GBS outperforms strong baselines, achieving a 9.8\% improvement in terms of MRR@20, over the state-of-the-art RIPOR method...

Figures

Figures reproduced from arXiv: 2501.11034 by the authors.

Figure 1
Figure 1. Books mainly consist of three types of information: [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Based on an encoder-decoder architecture, GBS comprises two components: (1) Data augmentation (orange dashed [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The performance, in terms of Hits@10, of GBS [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: The performance, in terms of Hits@10, of GBS [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

58 extracted references · 42 canonical work pages

  1. [1]

    Anserini

    Anserini 2020. Anserini. https://github.com/castorini/anserini

  2. [2]

    Federico Barrios, Federico López, Luis Argerich, and Rosa Wachenchauzer. 2016. Variations of the Similarity Function of TextRank for Automated Summarization. CoRR abs/1602.03606 (2016). arXiv:1602.03606 http://arxiv.org/abs/1602.03606

  3. [3]

    Michele Bevilacqua, Giuseppe Ottaviano, Patrick Lewis, Wen-tau Yih, Sebastian Riedel, and Fabio Petroni. 2022. Autoregressive Search Engines: Generating Substrings as Document Identifiers. In Advances in Neural Information Processing Systems. 31668–31683

  4. [4]

    Avi Bleiweiss. 2017. A Hierarchical Book Representation of Word Embeddings for Effective Semantic Clustering and Search. In International Conference on Agents and Artificial Intelligence, Vol. 2. SCITEPRESS, 154–163

  5. [5]

    Shubham Chatterjee and Laura Dietz. 2021. Entity Retrieval Using Fine-grained Entity Aspects. In proceedings of the 44th international ACM SIGIR conference on research and development in information retrieval . 1662–1666

  6. [6]

    Jiangui Chen, Ruqing Zhang, Jiafeng Guo, Yiqun Liu, Yixing Fan, and Xueqi Cheng. 2022. CorpusBrain: Pre-train a Generative Retrieval Model for Knowledge- Intensive Language Tasks. InProceedings of the 31st ACM International Conference on Information & Knowledge Management . 191–200

  7. [7]

    Djork-Arné Clevert, Thomas Unterthiner, and Sepp Hochreiter. 2015. Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs). arXiv preprint arXiv:1511.07289 (2015)

  8. [8]

    Arman Cohan, Franck Dernoncourt, Doo Soon Kim, Trung Bui, Seokhwan Kim, Walter Chang, and Nazli Goharian. 2018. A Discourse-aware Attention Model for Abstractive Summarization of Long Documents. arXiv preprint arXiv:1804.05685 (2018)

Show all 58 references
  1. [9]

    Nicola De Cao, Gautier Izacard, Sebastian Riedel, and Fabio Petroni. 2021. Autore- gressive Entity Retrieval. InInternational Conference on Learning Representations

  2. [10]

    Zican Dong, Tianyi Tang, Lunyi Li, and Wayne Xin Zhao. 2023. A Survey on Long Text Modeling with Transformers. arXiv preprint arXiv:2302.14502 (2023)

  3. [11]

    Skhumbuzo Dube and Ritesh Ajoodha. 2021. Improving Library Book Retrieval By Using Topic Modeling. In Interdisciplinary Research in Technology and Man- agement. CRC Press, 585–590

  4. [12]

    M Ali Fauzi, Agus Zainal Arifin, and Anny Yuniarti. 2017. Arabic Book Retrieval Using Class and Book Index Based Term Weighting. International Journal of Electrical and Computer Engineering 7, 6 (2017), 3705

  5. [13]

    Zhenyu He, Guhao Feng, Shengjie Luo, Kai Yang, Di He, Jingjing Xu, Zhi Zhang, Hongxia Yang, and Liwei Wang. 2024. Two Stones Hit One Bird: Bilevel Positional Encoding for Better Length Extrapolation. arXiv preprint arXiv:2401.16421 (2024)

  6. [14]

    Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, and Wu. 2020. Dense Passage Retrieval for Open-domain Question Answering. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing . 6769–6781

  7. [15]

    Angelos Katharopoulos, Apoorv Vyas, Nikolaos Pappas, and François Fleuret

  8. [16]

    Shah Khusro, Irfan Ullah, Azhar Rauf, and Saeed Mahfooz. 2014. Issues and Challenges in Book Information Retrieval. International Information Institute (Tokyo). Information 17, 6 (2014), 2055

  9. [17]

    Yongqi Li, Nan Yang, Liang Wang, Furu Wei, and Wenjie Li. 2023. Multiview Iden- tifiers Enhanced Generative Retrieval. In 61st Annual Meeting of the Association for Computational Linguistics. 6636–6648

  10. [18]

    Kevin Lin, Kyle Lo, Joseph E Gonzalez, and Dan Klein. 2023. Decomposing Complex Queries for Tip-of-the-tongue Retrieval. arXiv preprint arXiv:2305.15053 (2023)

  11. [19]

    Donald Metzler, Yi Tay, Dara Bahri, and Marc Najork. 2021. Rethinking Search: Making Domain Experts Out of Dilettantes. SIGIR Forum 55, 1 (2021), 1–27

  12. [20]

    Tsendsuren Munkhdalai, Manaal Faruqui, and Siddharth Gopal. 2024. Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention. arXiv preprint arXiv:2404.07143 (2024)

  13. [21]

    Tsendsuren Munkhdalai, Alessandro Sordoni, Tong Wang, and Adam Trischler

  14. [22]

    Ping Nie, Yuyu Zhang, Xiubo Geng, Arun Ramamurthy, Le Song, and Daxin Jiang

  15. [23]

    Rodrigo Nogueira and Jimmy Lin. 2019. From doc2query to docTTTTTquery. An MS MARCO Passage Retrieval Task Publication. University of Waterloo

  16. [24]

    Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. 2022. Training Language Models to Follow Instructions with Human Feedback. In Advances in Neural Information Processing Systems...

  17. [25]

    In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval

    DC-BERT: Decoupling Question and Document for Efficient Contextual Encoding. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval . ACM, 1829–1832

  18. [26]

    Julian Risch, Samuele Garda, and Ralf Krestel. 2018. Book Recommendation Beyond the Usual Suspects: Embedding Book Plots together with Place and Time Information. In Maturity and Innovation in Digital Libraries: 20th International Conference on Asia-Pacific Digital Libraries, ...

  19. [27]

    Robertson, Steve Walker, Susan Jones, Micheline M

    Stephen E. Robertson, Steve Walker, Susan Jones, Micheline M. Hancock-Beaulieu, and Mike Gatford. 1995. Okapi at TREC-3. In TREC. 109–126

  20. [28]

    Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. 2020. Exploring the Limits of Transfer Learning with a Unified Text-to-text Transformer.The Journal of Machine Learning Research 21, 1 (2020), 5485–5551

  21. [29]

    Imanol Schlag, Tsendsuren Munkhdalai, and Jürgen Schmidhuber. 2020. Learning Associative Inference using Fast Weight Memory.arXiv preprint arXiv:2011.07831 (2020)

  22. [30]

    Peter Shaw, Jakob Uszkoreit, and Ashish Vaswani. 2018. Self-attention with Relative Position Representations. arXiv preprint arXiv:1803.02155 (2018)

  23. [31]

    Anian Ruoss, Grégoire Delétang, Tim Genewein, Jordi Grau-Moya, Róbert Csor- dás, Mehdi Bennani, Shane Legg, and Joel Veness. 2023. Randomized Posi- tional Encodings Boost Length Generalization of Transformers. arXiv preprint arXiv:2305.16843 (2023)

  24. [32]

    Jianlin Su, Murtadha Ahmed, Yu Lu, Shengfeng Pan, Wen Bo, and Yunfeng Liu. 2024. Roformer: Enhanced Transformer with Rotary Position Embedding. Neurocomputing 568 (2024), 127063

  25. [33]

    Weiwei Sun, Lingyong Yan, Zheng Chen, Shuaiqiang Wang, Haichao Zhu, Pengjie Ren, Zhumin Chen, Dawei Yin, Maarten de Rijke, and Zhaochun Ren. 2023. Learning to Tokenize for Generative Retrieval. InAdvances in Neural Information Processing Systems, Vol. 36

  26. [34]

    Zhuoran Shen, Mingyuan Zhang, Haiyu Zhao, Shuai Yi, and Hongsheng Li. 2021. Efficient Attention: Attention with Linear Complexities. In Proceedings of the IEEE/CVF winter conference on applications of computer vision . 3531–3539

  27. [35]

    Yubao Tang, Ruqing Zhang, Jiafeng Guo, and Maarten de Rijke. 2023. Recent Advances in Generative Information Retrieval. In SIGIR-AP 2023: 1st International ACM SIGIR Conference on Information Retrieval in the Asia Pacific. ACM, 294–297

  28. [36]

    Tran, Mostafa Dehghani, Jianmo Ni, Dara Bahri, Harsh Mehta, Zhen Qin, Kai Hui, Zhe Zhao, Jai Gupta, Tal Schuster, William W

    Yi Tay, Vinh Q. Tran, Mostafa Dehghani, Jianmo Ni, Dara Bahri, Harsh Mehta, Zhen Qin, Kai Hui, Zhe Zhao, Jai Gupta, Tal Schuster, William W. Cohen, and Donald Metzler. 2022. Transformer Memory as a Differentiable Search Index. In Advances in Neural Information Processing Syste...

  29. [37]

    Yubao Tang, Ruqing Zhang, Jiafeng Guo, Jiangui Chen, Zuowei Zhu, Shuaiqiang Wang, Dawei Yin, and Xueqi Cheng. 2023. Semantic-Enhanced Differentiable Search Index Inspired by Learning Strategies. In 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . 4904–4913

  30. [38]

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is All You Need. Advances in neural information processing systems 30 (2017)

  31. [39]

    Yujing Wang, Yingyan Hou, Haonan Wang, Ziming Miao, Shibin Wu, Hao Sun, Qi Chen, Yuqing Xia, Chengmin Chi, Guoshuai Zhao, Zheng Liu, Xing Xie, Hao Sun, Weiwei Deng, Qi Zhang, and Mao Yang. 2022. A Neural Corpus Indexer for Document Retrieval. In Advances in Neural Information ...

  32. [40]

    Irfan Ullah and Shah Khusro. 2020. Social Book Search: The Impact of the Social Web on Book Retrieval and Recommendation. Multimedia Tools and Applications 79, 11 (2020), 8011–8060

  33. [41]

    Hengzhi Wu, Gabriella Kazai, and Michael Taylor. 2008. Book Search Experiments: Investigating IR Methods for the Indexing and Retrieval of Books. In Advances in Information Retrieval: 30th European Conference on IR Research, ECIR 2008, Glasgow, UK, March 30-April 3, 2008. Proc...

  34. [42]

    Haihua Xie, Zhiyou Chen, Yabo Li, Shaoling Jing, Xiaoqing Lyu, and Zhi Tang

  35. [43]

    Zihan Wang, Yujia Zhou, Yiteng Tu, and Zhicheng Dou. 2023. NOVO: Learnable and Interpretable Document Identifiers for Model-Based IR. In Proceedings of the 32nd ACM Conference on Information and Knowledge Management

  36. [44]

    Tianchi Yang, Minghui Song, Zihan Zhang, Haizhen Huang, Weiwei Deng, Feng Sun, and Qi Zhang. 2023. Auto Search Indexer for End-to-End Document Retrieval. In Findings of the Association for Computational Linguistics: EMNLP 2023

  37. [45]

    Zichao Yang, Diyi Yang, Chris Dyer, Xiaodong He, Alex Smola, and Eduard Hovy. 2016. Hierarchical Attention Networks for Document Classification. In Proceedings of the 2016 conference of the North American chapter of the association for computational linguistics: human language...

  38. [46]

    In 2020 IEEE conference on multimedia information processing and retrieval (MIPR)

    Sembrs: A Semantic Analysis Based Book Retrieval Approach. In 2020 IEEE conference on multimedia information processing and retrieval (MIPR) . IEEE, 101–104

  39. [47]

    Caiming Xiong, Victor Zhong, and Richard Socher. 2017. DCN+: Mixed Objec- tive And Deep Residual Coattention for Question Answering. In International Conference on Learning Representations

  40. [48]

    Jingtao Zhan, Jiaxin Mao, Yiqun Liu, Min Zhang, and Shaoping Ma. 2020. Rep- BERT: Contextualized Text Embeddings for First-stage Retrieval. arXiv preprint arXiv:2006.15498 (2020). Generative Retrieval for Book Search KDD ’25, August 3–7, 2025, Toronto, ON, Canada

  41. [49]

    Zhuosheng Zhang, Hanqing Zhang, Keming Chen, Yuhang Guo, Jingyun Hua, Yulong Wang, and Ming Zhou. 2021. Mengzi: Towards Lightweight yet Ingenious Pre-trained Models for Chinese. arXiv:2110.06696 [cs.CL]

  42. [50]

    Hansi Zeng, Chen Luo, Bowen Jin, Sheikh Muhammad Sarwar, Tianxin Wei, and Hamed Zamani. 2024. Scalable and Effective Generative Information Retrieval. In The 2024 ACM Web Conference

  43. [51]

    Hansi Zeng, Chen Luo, and Hamed Zamani. 2024. Planning Ahead in Generative Retrieval: Guiding Autoregressive Generation through Simultaneous Decoding. In Proceedings of the 2024 ACM SIGIR International Conference on Theory of Infor- mation Retrieval

  44. [52]

    Shengyao Zhuang, Houxing Ren, Linjun Shou, Jian Pei, Ming Gong, Guido Zuc- con, and Daxin Jiang. 2023. Bridging the Gap between Indexing and Retrieval for Differentiable Search Index with Query Generation. In Gen-IR@SIGIR 2023: The First Workshop on Generative Information Retr...

  45. [54]

    Yujia Zhou, Jing Yao, Zhicheng Dou, Ledell Wu, Peitian Zhang, and Ji-Rong Wen

  46. [56]

    Dawei Zhu, Nan Yang, Liang Wang, Yifan Song, Wenhao Wu, Furu Wei, and Sujian Li. 2023. Pose: Efficient Context Window Extension of LLMs via Positional Skip-wise Training. arXiv preprint arXiv:2309.10400 (2023)

  47. [58]

    The Heart of a Boy

    uses an autoencoder to generate identifiers for books, which compress book contents into identifiers and to reconstruct docids back into book contents. It learns jointly with the retrieval task. (ix) NOVO [40] selects important words from the book as identifiers. The model is ...

  48. [2019]

    Advances in Neural Information Processing Systems 32 (2019)

    Metalearned Neural Memory. Advances in Neural Information Processing Systems 32 (2019)

  49. [2020]

    In International Conference on Machine Learning

    Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention. In International Conference on Machine Learning . PMLR, 5156–5165

  50. [2022]

    Ultron: An Ultimate Retriever on Corpus with a Model-based Indexer.arXiv preprint arXiv:2208.09257 (2022)

Pith tools

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