pith. machine review for the scientific record. sign in

arxiv: 2604.07825 · v2 · submitted 2026-04-09 · 💻 cs.IR · cs.AI

Recognition: unknown

Filling the Gaps: Selective Knowledge Augmentation for LLM Recommenders

Jaehyun Lee , Sanghwan Jang , SeongKu Kang , HwanJo Yu
Authors on Pith no claims yet

Pith reviewed 2026-05-10 17:31 UTC · model grok-4.3

classification 💻 cs.IR cs.AI
keywords LLM recommendersknowledge gapsselective augmentationcollaborative probingcontext efficiencytraining-free recommendationitem knowledge
0
0 comments X

The pith

Selective probing identifies LLM knowledge gaps so external information can be added only where needed for better recommendations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Large language models acting as training-free recommenders suffer from uneven item knowledge caused by imbalanced pretraining data. Uniform methods that append external details to every item in the prompt waste context space on items the model already understands and can even impair its reasoning. KnowSA_CKP first probes the model by checking how well it captures collaborative relationships among items, then augments only the items where this probing signals a gap. This focused use of the context budget raises both accuracy and efficiency on four real-world datasets without any fine-tuning step.

Core claim

The paper establishes that an LLM's internal knowledge of items can be estimated through comparative probing of its ability to model collaborative patterns, allowing external information to be injected selectively rather than uniformly; this targeted supplementation improves recommendation performance by concentrating limited context resources on the items that most need it.

What carries the argument

KnowSA_CKP, which performs comparative knowledge probing to rank items by estimated knowledge gaps and then applies selective augmentation only to high-gap items.

If this is right

  • Recommendation accuracy rises while total tokens in the prompt decrease because well-known items receive no extra text.
  • The model avoids distraction from redundant facts and maintains stronger reasoning over the user history.
  • The same probing-plus-selective pattern works across multiple datasets without requiring model retraining or parameter updates.
  • Context budget is freed for longer user histories or more candidate items within the same token limit.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The probing technique could be adapted to detect gaps in other LLM tasks such as open-ended question answering or summarization.
  • Combining the method with dynamic context allocation might further reduce costs in production recommendation systems.
  • If the probe can be made cheaper, the approach scales to very large item catalogs where uniform augmentation is prohibitive.

Load-bearing premise

The model's performance on a collaborative-relationship probing task accurately reflects its true gaps in item-specific knowledge.

What would settle it

A controlled test in which items flagged as knowledge-gap items by the probe show no larger accuracy lift from augmentation than randomly selected items, or in which different probing tasks yield inconsistent gap rankings.

Figures

Figures reproduced from arXiv: 2604.07825 by HwanJo Yu, Jaehyun Lee, Sanghwan Jang, SeongKu Kang.

Figure 1
Figure 1. Figure 1: Recommendation performance change with uni [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Alignment between knowledge scores and recommendation quality on the A-Beauty dataset. (a) Spearman correlation [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Average Recall@1 across four quantile bins grouped [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: Effect of 𝐾aug (left) and 𝐾ref (right) in KnowSACKP [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
read the original abstract

Large language models (LLMs) have recently emerged as powerful training-free recommenders. However, their knowledge of individual items is inevitably uneven due to imbalanced information exposure during pretraining, a phenomenon we refer to as knowledge gap problem. To address this, most prior methods have employed a naive uniform augmentation that appends external information for every item in the input prompt. However, this approach not only wastes limited context budget on redundant augmentation for well-known items but can also hinder the model's effective reasoning. To this end, we propose KnowSA_CKP (Knowledge-aware Selective Augmentation with Comparative Knowledge Probing) to mitigate the knowledge gap problem. KnowSA_CKP estimates the LLM's internal knowledge by evaluating its capability to capture collaborative relationships and selectively injects additional information only where it is most needed. By avoiding unnecessary augmentation for well-known items, KnowSA_CKP focuses on items that benefit most from knowledge supplementation, thereby making more effective use of the context budget. KnowSA_CKP requires no fine-tuning step, and consistently improves both recommendation accuracy and context efficiency across four real-world datasets. Our code is available at https://github.com/nowhyun/KnowSA\_CKP.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper introduces KnowSA_CKP, a training-free method for LLM-based recommenders to address uneven item knowledge ('knowledge gap problem') from pretraining imbalances. Rather than uniform external augmentation for all items (which wastes context and may impair reasoning), it employs Comparative Knowledge Probing to evaluate the LLM's ability to capture collaborative signals from user history, estimates per-item knowledge gaps, and selectively augments only the most needed items. The approach claims consistent gains in recommendation accuracy and context efficiency across four real-world datasets, with no fine-tuning and public code.

Significance. If the probing mechanism reliably isolates true item-level knowledge gaps, the selective augmentation could meaningfully improve context utilization and performance for LLM recommenders without training costs. Public code and training-free design are clear strengths that support reproducibility. However, the result's significance depends on whether the empirical gains hold under rigorous controls and whether the probing assumption is validated rather than assumed.

major comments (2)
  1. [§3 (KnowSA_CKP method and probing description)] The central assumption that comparative knowledge probing success directly measures the LLM's internal item knowledge (rather than reasoning ability, prompt sensitivity, or dataset-specific patterns) is load-bearing for the selective-injection claim but receives insufficient validation. If probing failures arise from non-knowledge factors, the method may augment the wrong items or skip real gaps, weakening both accuracy and efficiency results.
  2. [§4 (Experiments)] The experimental section reports consistent improvements on four datasets but provides no details on exact baselines, metrics, statistical significance tests, or controls for prompt variations and augmentation volume. Without these, it is impossible to assess whether the gains are attributable to selective augmentation or other factors.
minor comments (2)
  1. [Abstract] The abstract would benefit from naming the specific datasets, metrics, and magnitude of improvements to allow readers to gauge the claims immediately.
  2. [§3] Notation for the probing score and augmentation threshold should be defined more explicitly with an equation or pseudocode for clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their insightful comments, which have helped us identify areas for improvement in the manuscript. We address each major comment below and indicate the revisions we plan to make.

read point-by-point responses

  1. Referee: The central assumption that comparative knowledge probing success directly measures the LLM's internal item knowledge (rather than reasoning ability, prompt sensitivity, or dataset-specific patterns) is load-bearing for the selective-injection claim but receives insufficient validation. If probing failures arise from non-knowledge factors, the method may augment the wrong items or skip real gaps, weakening both accuracy and efficiency results.

    Authors: We appreciate this concern regarding the validation of the probing mechanism. The design of Comparative Knowledge Probing is intended to assess the LLM's ability to leverage collaborative signals from user history for item prediction, which we argue serves as a proxy for internal knowledge of the item. To provide stronger evidence, we will include additional experiments in the revised manuscript, such as correlating probing scores with item popularity (as a proxy for pretraining exposure) and testing probing robustness across different prompt phrasings. This will help isolate knowledge gaps from other factors. revision: yes

  2. Referee: The experimental section reports consistent improvements on four datasets but provides no details on exact baselines, metrics, statistical significance tests, or controls for prompt variations and augmentation volume. Without these, it is impossible to assess whether the gains are attributable to selective augmentation or other factors.

    Authors: We regret that the experimental details were not sufficiently clear in the initial submission. We will revise the experimental section to explicitly detail the baselines used, the evaluation metrics, the statistical significance tests performed, and include additional controls for prompt variations and augmentation volume to better attribute the observed gains to the selective augmentation approach. revision: yes

Circularity Check

0 steps flagged

Empirical method with no derivation chain or self-referential reductions

full rationale

The paper introduces KnowSA_CKP as a practical, training-free algorithm that uses comparative knowledge probing to decide selective augmentation for LLM recommenders. It contains no equations, first-principles derivations, or mathematical predictions that could reduce to fitted parameters or self-definitions. Claims rest on experimental results across four datasets and public code, with no load-bearing self-citations or ansatzes that collapse the method to its inputs. This is self-contained empirical work; the central mechanism (probing success as proxy for knowledge gaps) is an explicit design choice, not a hidden equivalence.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Based on abstract only; no free parameters, axioms, or invented entities are described. The method uses standard LLM prompting and probing without introducing new mathematical constructs or entities.

pith-pipeline@v0.9.0 · 5516 in / 953 out tokens · 85293 ms · 2026-05-10T17:31:13.462243+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

65 extracted references · 26 canonical work pages · 9 internal anchors

  1. [1]

    Davide Abbattista, Vito Walter Anelli, Tommaso Di Noia, Craig Macdonald, and Aleksandr Vladimirovich Petrov. 2024. Enhancing sequential music recommen- dation with personalized popularity awareness. InProceedings of the 18th ACM Conference on Recommender Systems. 1168–1173

    2024

  2. [2]

    Himan Abdollahpouri, Robin Burke, and Bamshad Mobasher. 2019. Managing popularity bias in recommender systems with personalized re-ranking.arXiv preprint arXiv:1901.07555(2019)

    work page arXiv 2019

  3. [3]

    Alfonso Amayuelas, Kyle Wong, Liangming Pan, Wenhu Chen, and William Wang. 2023. Knowledge of knowledge: Exploring known-unknowns uncertainty with large language models.arXiv preprint arXiv:2305.13712(2023)

    work page arXiv 2023

  4. [4]

    Keqin Bao, Jizhi Zhang, Wenjie Wang, Yang Zhang, Zhengyi Yang, Yanchen Luo, Chong Chen, Fuli Feng, and Qi Tian. 2025. A bi-step grounding paradigm for large language models in recommendation systems.ACM Transactions on Recommender Systems3, 4 (2025), 1–27

    2025

  5. [5]

    Keqin Bao, Jizhi Zhang, Yang Zhang, Wenjie Wang, Fuli Feng, and Xiangnan He. 2023. Tallrec: An effective and efficient tuning framework to align large language model with recommendation. InProceedings of the 17th ACM Conference on Recommender Systems. 1007–1014

    2023

  6. [6]

    Zhe Cao, Tao Qin, Tie-Yan Liu, Ming-Feng Tsai, and Hang Li. 2007. Learning to rank: from pairwise approach to listwise approach. InProceedings of the 24th international conference on Machine learning. 129–136

    2007

  7. [7]

    Nicholas Carlini, Florian Tramer, Eric Wallace, Matthew Jagielski, Ariel Herbert- Voss, Katherine Lee, Adam Roberts, Tom Brown, Dawn Song, Ulfar Erlingsson, et al. 2021. Extracting training data from large language models. In30th USENIX security symposium (USENIX Security 21). 2633–2650

    2021

  8. [8]

    Jiaju Chen, Chongming Gao, Shuai Yuan, Shuchang Liu, Qingpeng Cai, and Peng Jiang. 2025. Dlcrec: A novel approach for managing diversity in llm- based recommender systems. InProceedings of the Eighteenth ACM International Conference on Web Search and Data Mining. 857–865

    2025

  9. [9]

    Paul Covington, Jay Adams, and Emre Sargin. 2016. Deep neural networks for youtube recommendations. InProceedings of the 10th ACM conference on recommender systems. 191–198

    2016

  10. [10]

    Jinhao Duan, Hao Cheng, Shiqi Wang, Alex Zavalny, Chenan Wang, Renjing Xu, Bhavya Kailkhura, and Kaidi Xu. 2024. Shifting attention to relevance: Towards the predictive uncertainty quantification of free-form large language models. InProceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 5050–5063

    2024

  11. [11]

    Ekaterina Fadeeva, Aleksandr Rubashevskii, Artem Shelmanov, Sergey Petrakov, Haonan Li, Hamdy Mubarak, Evgenii Tsymbalov, Gleb Kuzmin, Alexander Panchenko, Timothy Baldwin, et al . 2024. Fact-checking the output of large language models via token-level uncertainty quantification.arXiv preprint arXiv:2403.04696(2024)

    work page arXiv 2024

  12. [12]

    Marina Fomicheva, Shuo Sun, Lisa Yankovskaya, Frédéric Blain, Francisco Guzmán, Mark Fishel, Nikolaos Aletras, Vishrav Chaudhary, and Lucia Specia

  13. [13]

    Unsupervised quality estimation for neural machine translation.Transac- tions of the Association for Computational Linguistics8 (2020), 539–555

    2020

  14. [14]

    Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, et al. 2024. The llama 3 herd of models.arXiv preprint arXiv:2407.21783 (2024)

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  15. [15]

    F Maxwell Harper and Joseph A Konstan. 2015. The movielens datasets: History and context.Acm transactions on interactive intelligent systems (tiis)5, 4 (2015), 1–19

    2015

  16. [16]

    Jesse Harte, Wouter Zorgdrager, Panos Louridas, Asterios Katsifodimos, Diet- mar Jannach, and Marios Fragkoulis. 2023. Leveraging large language models for sequential recommendation. InProceedings of the 17th ACM Conference on Recommender Systems. 1096–1102

    2023

  17. [17]

    Yupeng Hou, Junjie Zhang, Zihan Lin, Hongyu Lu, Ruobing Xie, Julian McAuley, and Wayne Xin Zhao. 2024. Large language models are zero-shot rankers for recommender systems. InEuropean Conference on Information Retrieval. Springer, 364–381. Filling the Gaps: Selective Knowledge Augmentation for LLM Recommenders SIGIR ’26, July 20–24, 2026, Melbourne, VIC, Australia

    2024

  18. [18]

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen, et al. 2022. Lora: Low-rank adaptation of large language models.ICLR1, 2 (2022), 3

    2022

  19. [19]

    Aaron Hurst, Adam Lerer, Adam P Goucher, Adam Perelman, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Welihinda, Alan Hayes, Alec Radford, et al. 2024. Gpt-4o system card.arXiv preprint arXiv:2410.21276(2024)

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  20. [20]

    Soyeong Jeong, Jinheon Baek, Sukmin Cho, Sung Ju Hwang, and Jong Park

  21. [21]

    Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity. InProceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), Kevin Duh, Helena Gomez, and Steven Bethard (Eds.). Association for Computational Li...

    work page doi:10.18653/v1/2024.naacl-long.389 2024

  22. [22]

    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, Lélio Renard Lavaud, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, and William El Sayed. 2023. Mistral 7B.ArXivabs/2310.0...

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  23. [23]

    Zhengbao Jiang, Frank F Xu, Luyu Gao, Zhiqing Sun, Qian Liu, Jane Dwivedi-Yu, Yiming Yang, Jamie Callan, and Graham Neubig. 2023. Active retrieval augmented generation. InProceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 7969–7992

    2023

  24. [24]

    Cheongwoong Kang and Jaesik Choi. 2023. Impact of co-occurrence on factual knowledge of large language models.arXiv preprint arXiv:2310.08256(2023)

    work page arXiv 2023

  25. [25]

    Wang-Cheng Kang and Julian McAuley. 2018. Self-attentive sequential recom- mendation. In2018 IEEE international conference on data mining (ICDM). IEEE, 197–206

    2018

  26. [26]

    Wang-Cheng Kang and Julian McAuley. 2019. Candidate generation with binary codes for large-scale top-n recommendation. InProceedings of the 28th ACM international conference on information and knowledge management. 1523–1532

    2019

  27. [27]

    Jieyong Kim, Hyunseo Kim, Hyunjin Cho, SeongKu Kang, Buru Chang, Jinyoung Yeo, and Dongha Lee. 2025. Review-driven Personalized Preference Reasoning with Large Language Models for Recommendation. InProceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR ’25). Association for Computing Machiner...

    2025

  28. [28]

    Sein Kim, Hongseok Kang, Seungyoon Choi, Donghyun Kim, Minchul Yang, and Chanyoung Park. 2024. Large language models meet collaborative filtering: An efficient all-round llm-based recommender system. InProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 1395–1406

    2024

  29. [29]

    Sunwoo Kim, Geon Lee, Kyungho Kim, Jaemin Yoo, and Kijung Shin. 2025. Item- RAG: Item-Based Retrieval-Augmented Generation for LLM-Based Recommen- dation.arXiv preprint arXiv:2511.15141(2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  30. [30]

    Yueqing Liang, Liangwei Yang, Chen Wang, Xiongxiao Xu, Philip S Yu, and Kai Shu. 2024. Taxonomy-Guided Zero-Shot Recommendations with LLMs.arXiv preprint arXiv:2406.14043(2024)

    work page arXiv 2024

  31. [31]

    Jiayi Liao, Sihang Li, Zhengyi Yang, Jiancan Wu, Yancheng Yuan, and Xiang Wang

  32. [32]

    Llara: Aligning large language models with sequential recommenders.CoRR (2023)

    2023

  33. [33]

    Jianghao Lin, Rong Shan, Chenxu Zhu, Kounianhua Du, Bo Chen, Shigang Quan, Ruiming Tang, Yong Yu, and Weinan Zhang. 2024. Rella: Retrieval-enhanced large language models for lifelong sequential behavior comprehension in recom- mendation. InProceedings of the ACM Web Conference 2024. 3497–3508

    2024

  34. [34]

    Xinyu Lin, Wenjie Wang, Yongqi Li, Fuli Feng, See-Kiong Ng, and Tat-Seng Chua

  35. [35]

    InProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

    Bridging items and language: A transition paradigm for large language model-based recommendation. InProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 1816–1826

  36. [36]

    Xinyu Lin, Wenjie Wang, Yongqi Li, Shuo Yang, Fuli Feng, Yinwei Wei, and Tat- Seng Chua. 2024. Data-efficient Fine-tuning for LLM-based Recommendation. InProceedings of the 47th international ACM SIGIR conference on research and development in information retrieval. 365–374

    2024

  37. [37]

    Zhen Lin, Shubhendu Trivedi, and Jimeng Sun. 2024. Generating with Confi- dence: Uncertainty Quantification for Black-box Large Language Models.Trans- actions on Machine Learning Research(2024). https://openreview.net/forum?id= DWkJCSxKU5

    2024

  38. [38]

    Nelson F Liu, Kevin Lin, John Hewitt, Ashwin Paranjape, Michele Bevilacqua, Fabio Petroni, and Percy Liang. 2023. Lost in the middle: How language models use long contexts.arXiv preprint arXiv:2307.03172(2023)

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  39. [39]

    Qidong Liu, Xian Wu, Yejing Wang, Zijian Zhang, Feng Tian, Yefeng Zheng, and Xiangyu Zhao. 2024. Llm-esr: Large language models enhancement for long- tailed sequential recommendation.Advances in Neural Information Processing Systems37 (2024), 26701–26727

    2024

  40. [40]

    Sichun Luo, Bowei He, Haohan Zhao, Wei Shao, Yanlin Qi, Yinya Huang, Aojun Zhou, Yuxuan Yao, Zongpeng Li, Yuanzhang Xiao, et al. 2024. Recranker: Instruc- tion tuning large language model as ranker for top-k recommendation.ACM Transactions on Information Systems(2024)

    2024

  41. [41]

    Alex Mallen, Akari Asai, Victor Zhong, Rajarshi Das, Daniel Khashabi, and Hannaneh Hajishirzi. 2022. When not to trust language models: Investigat- ing effectiveness of parametric and non-parametric memories.arXiv preprint arXiv:2212.10511(2022)

    work page arXiv 2022

  42. [42]

    Jianmo Ni, Jiacheng Li, and Julian McAuley. 2019. Justifying recommendations using distantly-labeled reviews and fine-grained aspects. InProceedings of the 2019 conference on empirical methods in natural language processing and the 9th international joint conference on natural language processing (EMNLP-IJCNLP). 188–197

    2019

  43. [43]

    Aleksandr Petrov and Craig Macdonald. 2024. RSS: effective and efficient training for sequential recommendation using recency sampling.ACM Transactions on Recommender Systems3, 1 (2024), 1–32

    2024

  44. [44]

    Nils Reimers and Iryna Gurevych. 2019. Sentence-bert: Sentence embeddings using siamese bert-networks.arXiv preprint arXiv:1908.10084(2019)

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  45. [45]

    Xubin Ren, Wei Wei, Lianghao Xia, Lixin Su, Suqi Cheng, Junfeng Wang, Dawei Yin, and Chao Huang. 2024. Representation learning with large language models for recommendation. InProceedings of the ACM Web Conference 2024. 3464–3475

    2024

  46. [46]

    Steffen Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars Schmidt-Thieme

  47. [47]

    BPR: Bayesian personalized ranking from implicit feedback.arXiv preprint arXiv:1205.2618(2012)

    work page internal anchor Pith review arXiv 2012

  48. [48]

    Weijia Shi, Anirudh Ajith, Mengzhou Xia, Yangsibo Huang, Daogao Liu, Terra Blevins, Danqi Chen, and Luke Zettlemoyer. 2023. Detecting pretraining data from large language models.arXiv preprint arXiv:2310.16789(2023)

    work page arXiv 2023

  49. [49]

    Weihang Su, Yichen Tang, Qingyao Ai, Zhijing Wu, and Yiqun Liu. 2024. DRAGIN: Dynamic Retrieval Augmented Generation based on the Real-time Information Needs of Large Language Models. InProceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Lun-Wei Ku, Andre Martins, and Vivek Srikumar (Eds.). Asso...

    work page doi:10.18653/v1/2024.acl-long.702 2024

  50. [50]

    Qwen Team. 2024. Qwen2.5 Technical Report.ArXivabs/2412.15115 (2024)

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  51. [51]

    Shijie Wang, Wenqi Fan, Yue Feng, Xinyu Ma, Shuaiqiang Wang, and Dawei Yin. 2025. Knowledge Graph Retrieval-Augmented Generation for LLM-based Recommendation.arXiv preprint arXiv:2501.02226(2025)

    work page arXiv 2025

  52. [52]

    Jason Wei, Maarten Bosma, Vincent Y Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M Dai, and Quoc V Le. 2021. Finetuned language models are zero-shot learners.arXiv preprint arXiv:2109.01652(2021)

    work page internal anchor Pith review arXiv 2021

  53. [53]

    Zhiqiu Xia, Jinxuan Xu, Yuqian Zhang, and Hang Liu. 2025. A survey of uncertainty estimation methods on large language models.arXiv preprint arXiv:2503.00172(2025)

    work page arXiv 2025

  54. [54]

    Haoran Yang, Yumeng Zhang, Jiaqi Xu, Hongyuan Lu, Pheng Ann Heng, and Wai Lam. 2024. Unveiling the generalization power of fine-tuned large language models.arXiv preprint arXiv:2403.09162(2024)

    work page arXiv 2024

  55. [55]

    Yuqing Yang, Lei Jiao, and Yuedong Xu. 2024. A queueing theoretic perspective on low-latency llm inference with variable token length. In2024 22nd Interna- tional Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt). IEEE, 273–280

    2024

  56. [56]

    Zijun Yao, Weijian Qi, Liangming Pan, Shulin Cao, Linmei Hu, Liu Weichuan, Lei Hou, and Juanzi Li. 2025. SeaKR: Self-aware Knowledge Retrieval for Adaptive Retrieval Augmented Generation. InProceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics, Vienna, Aus...

    work page doi:10.18653/v1/ 2025

  57. [57]

    Qinan Yu, Jack Merullo, and Ellie Pavlick. 2023. Characterizing mechanisms for factual recall in language models.arXiv preprint arXiv:2310.15910(2023)

    work page arXiv 2023

  58. [58]

    Zhenrui Yue, Sara Rabhi, Gabriel de Souza Pereira Moreira, Dong Wang, and Even Oldridge. 2023. Llamarec: Two-stage recommendation using large language models for ranking.arXiv preprint arXiv:2311.02089(2023)

    work page arXiv 2023

  59. [59]

    Weichao Zhang, Ruqing Zhang, Jiafeng Guo, Maarten de Rijke, Yixing Fan, and Xueqi Cheng. 2024. Pretraining data detection for large language models: A divergence-based calibration method.arXiv preprint arXiv:2409.14781(2024)

    work page arXiv 2024

  60. [60]

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

  61. [61]

    Text-like encoding of collaborative information in large language models for recommendation.arXiv preprint arXiv:2406.03210(2024)

    work page arXiv 2024

  62. [62]

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

  63. [63]

    Collm: Integrating collaborative embeddings into large language models for recommendation.IEEE Transactions on Knowledge and Data Engineering(2025)

    2025

  64. [64]

    Zhi Zheng, Wenshuo Chao, Zhaopeng Qiu, Hengshu Zhu, and Hui Xiong. 2024. Harnessing large language models for text-rich sequential recommendation. In Proceedings of the ACM Web Conference 2024. 3207–3216

    2024

  65. [65]

    Baohang Zhou, Zezhong Wang, Lingzhi Wang, Hongru Wang, Ying Zhang, Kehui Song, Xuhui Sui, and Kam-Fai Wong. 2024. DPDLLM: A Black-box Framework for Detecting Pre-training Data from Large Language Models. InFindings of the Association for Computational Linguistics ACL 2024. 644–653

    2024