Pith. sign in

REVIEW 2 major objections 5 minor 75 references

LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation

T0 review · 2 major / 5 minor · reviewed 2026-07-11 · grok-4.5

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

Pith's one-line read LLM recommenders favor long or short titles by design; LBR removes that length bias on both attention and decoding.

desk verdict Solid systems paper that cleanly diagnoses dual length bias in LLM recommenders and ships a cheap, effective fix with real gains and code. read the letter →

arxiv 2607.04270 v1 pith:IQMBLYPA submitted 2026-07-05 cs.IR cs.AI

Hongchen Li , Bohao Wang , Jingbang Chen , Weiqin Yang , Hang Pan , Bingde Hu , Can Wang , Jiawei Chen This is my paper
classification cs.IRcs.AI
keywords SequentialRecommendationLargeLanguageModelsLengthBiasAttentionCalibrationConstrainedDecodingTrieFairness
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

When large language models recommend products from text titles, item length quietly warps the answer. Longer titles grab more attention mass in the model’s context, so they influence preference modeling more than they deserve; when the model scores candidates by summing log-probabilities, short titles win simply because fewer negative terms are added. Ordinary length normalization flips the bias the other way, because under a Trie of valid catalog tokens many positions are nearly deterministic and therefore cheap to predict. The paper’s claim is that both distortions can be cancelled with a lightweight, model-agnostic fix called LBR: a length-dependent offset inside the attention logits, plus a new score that normalizes by the total branching uncertainty along each item’s path rather than by raw token count. On three Amazon catalogs and two LLM recommenders the method raises ranking metrics by roughly seventeen percent on average while making exposure across length groups far more even, at almost no extra training or inference cost.

What carries the argument

LBR: Length-Aware Attention Calibration injects a learned offset −log(g(l)) into attention logits so cumulative attention becomes roughly length-invariant; Effective Information Length Normalization replaces token count by the sum of Hartley entropies log2|V_k| along the Trie path, then weights each log-probability by that same informativeness before averaging.

What would settle it

Re-run the attention-mass versus length plots (Figure 3 / Figure 6) after calibration; if a clear positive slope remains, or if NDCG@5 gains disappear once items are forced to equal token length, the central claim fails.

Watch

Extended reading notes

Core claim

Length bias is a structural artifact of token-level LLM recommendation: longer item texts accumulate excess attention on the input side, while summed log-likelihood (and naive length normalization) systematically favor short or long items on the output side. LBR removes both effects by Length-Aware Attention Calibration and Effective Information Length Normalization, producing higher accuracy and fairer length exposure with negligible overhead.

Load-bearing premise

The expected attention mass an item receives before calibration is assumed to be a simple multiplicative function of its length, so subtracting the log of that function cancels the bias.

Editorial extensions

If this is right

  • Any Transformer recommender that concatenates free-text item titles can plug in the same two corrections without changing architecture.
  • Catalog exposure becomes closer to the true length distribution in the data, reducing pressure on sellers to invent unnaturally short or padded titles.
  • Conventional length-penalty hyperparameters become unnecessary once branching entropy is used as the normalizer.
  • Fairness audits of LLM recommenders should report length-group variance alongside popularity and position metrics.

Reading between the lines

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

  • The same branching-entropy normalizer could improve constrained generation outside recommendation (e.g., product-attribute completion or catalog-constrained dialogue).
  • If the length–attention relationship turns out non-linear on other domains, a richer parametric g(l) could be swapped in without altering the rest of LBR.
  • Jointly modeling length bias with known popularity and position biases may yield further gains once the two axes are no longer confounded.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The paper identifies length bias in LLM-based sequential recommenders that represent items by variable-length text: longer items accumulate more attention mass on the input side, while summed autoregressive log-likelihood on the output side (under Trie-constrained beam search) systematically favors short items; standard length normalization can reverse the bias and hurt accuracy. It proposes LBR, a lightweight model-agnostic fix consisting of (i) Length-Aware Attention Calibration (LAAC), which adds a learned length-dependent offset δ(l_v)=−log(g(l_v)) with linear g(l)=al+b into attention logits, and (ii) Effective Information Length Normalization (EILN), which replaces raw token count by a Hartley-entropy surrogate U_y=∑log2|V_k| derived from Trie branching factors and reweights token log-probabilities accordingly. Experiments on three Amazon datasets (Toy, Office, Book) with BIGRec and LLaRA backbones report average NDCG@5 gains of ~16.8% over strong baselines, ablations isolating both components, attention-weight and length-group exposure plots, and negligible overhead; code is released.

Significance. If the dual-bias diagnosis and the reported gains hold, the work is a useful systems contribution for the growing class of generative LLM recommenders that decode under catalog Tries. The problem is concrete and previously under-addressed; the remedies are simple, model-agnostic, and cheap (two extra scalars plus O(N) bookkeeping). Strengths include consistent gains across three datasets and two backbones, component ablations (Table 3), direct evidence that post-calibration attention becomes nearly length-invariant (Fig. 6) and that exposure variance across length groups shrinks (Fig. 7), statistical-significance claims, and public code. The information-theoretic length surrogate is a clean, falsifiable design choice that improves on both unnormalized scoring and classical length penalty. The contribution is primarily empirical/systems rather than theoretical, but that is appropriate for the venue and claim.

major comments (2)
  1. §4.1 and Appendix F, Lemma 1 / Assumption (A2): the theoretical claim that calibrated expected attention is length-invariant rests on a multiplicative pre-calibration model E[Z_u(i)|l_i=l]=κ_u g(l) and a concentration approximation for the softmax normalizer. The paper already notes that g is learned end-to-end under the recommendation loss (not fitted to attention) and that non-linear alternatives give negligible gains; Fig. 6 further shows the practical effect. The lemma is therefore only an approximate justification. Either weaken the claim to “motivating analysis under mild assumptions” or add a short empirical check that the learned g recovers a roughly multiplicative trend on held-out attention mass, so the theory and practice stay aligned.
  2. §5.2 / Table 1 and §5.4 / Fig. 7: fairness is asserted via reduced variance of recommendation proportions across length groups that are equalized by cumulative item frequency. This is a reasonable proxy, but it is not a standard item-side fairness metric (e.g., exposure parity conditioned on popularity or provider-level metrics). A brief additional analysis that the accuracy gains are not driven by a reallocation that harms already under-exposed long/short items (or a simple popularity-controlled length-group breakdown) would make the fairness claim more robust without changing the experimental design.
minor comments (5)
  1. Eq. (9) is labeled s_EIF while the surrounding text uses EIL / EILN; unify the acronym (EILN vs EIF) for consistency.
  2. Fig. 4 caption and §3.2: “number of valid tokens” is clear, but a short note that |V_k| is the Trie branching factor at step k would help readers who skip the figure.
  3. Appendix D (equal-length truncation/padding and RQ-VAE semantic IDs) is useful negative evidence; a one-sentence pointer in the main §3.1 would make the “naive remedies fail” claim easier to verify.
  4. Table 1 improvement percentages are relative to the best baseline per column; stating this explicitly in the caption would avoid ambiguity when the best baseline changes across metrics.
  5. Minor typography: “Length-A ware” (space) appears in the abstract and §4.1; “seemingly integrated” in §5.2 should be “seamlessly”.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: LAAC offset and EILN are design choices evaluated on external metrics, not predictions forced by construction or self-citation.

full rationale

The paper's central claims (accuracy/fairness gains from dual length-bias mitigation) rest on empirical evaluation with NDCG@K, Hit@K, and length-group exposure variance against test-set ideal distributions, all external to the proposed normalizers. LAAC introduces a length-dependent logit offset δ(l_v) = -log(g(l_v)) with g(l) = al + b; a and b are two free parameters optimized jointly end-to-end under the primary recommendation loss (Section 4.1), not fitted to the attention-length scatter of Figure 3 and then re-used as a 'prediction' of invariance. Lemma 1 (Appendix F) merely shows that, under the mild modeling assumptions (A1 concentration, A2 multiplicative length trend matching the observed linear pattern), the calibrated expected attribution becomes approximately length-invariant; this is a standard cancellation argument for the chosen functional form, not a self-definitional derivation of the method from the target result. EILN replaces raw length by the sum of Hartley entropies log2|V_k| taken from the Trie branching factors (Eqs. 7-9); the formula is fixed by the decoding constraint structure and is not fitted to observed recommendation frequencies. No uniqueness theorem, load-bearing self-citation chain, or renaming of a known empirical pattern appears. Ablations (Table 3), attention plots (Figure 6), and decoding histograms (Figure 7) provide independent checks. The derivation chain is therefore self-contained and non-circular.

Assumptions & free parameters 1 free parameters · 3 assumptions · 2 invented entities

The central empirical claim rests on two lightweight modeling choices (linear length–attention trend and Hartley entropy of Trie branch width) plus standard Transformer and constrained-beam-search machinery. Only two scalar parameters are learned; no new physical entities are postulated.

free parameters (1)
  • a, b of g(l)=a l + b
    Linear coefficients of the length-to-attention-mass function; jointly optimized with the recommendation objective rather than fixed a priori.
assumptions (3)
  • domain assumption Self-attention logits can be additively offset by a length-dependent term without destroying the model's ability to model user preference.
    Invoked in §4.1 when defining A'_uv; standard in attention-calibration literature but not proved for recommendation prompts.
  • ad hoc to paper Token informativeness under Trie-constrained decoding is well-approximated by Hartley entropy log2 |V_k|.
    Core modeling choice of EILN (§4.2); motivated by information theory but not derived from first principles of the LLM likelihood.
  • ad hoc to paper Assumption (A1)–(A2) of Lemma 1: softmax normalizer concentrates and pre-calibration attention mass is multiplicative in length.
    Required for the length-invariance proof in Appendix F; supported by the observed linear trend in Fig. 3 but not independently verified.
invented entities (2)
  • Effective Information Length (EIL / U_y)
    purpose: Replace raw token count with a branching-aware length surrogate for score normalization.
    Defined as sum of log2 of valid-token set sizes along the Trie path; no external measurement of 'information length' is supplied.
  • Length-Aware Attention Calibration offset δ(l_v)=−log(g(l_v))
    purpose: Cancel the multiplicative length advantage inside softmax attention.
    Constructed specifically for this paper from the fitted g(l); not previously standard in recsys literature.

how reviews work

0 comments
Cite this review

Pith. "Pith review of LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation." pith.science (2026). https://pith.science/paper/IQMBLYPA

@misc{pith2026260704270,
  author       = {Pith},
  title        = {Pith review of: LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IQMBLYPA}},
  note         = {Machine review of arXiv:2607.04270}
}
abstract

Large language models (LLMs) have recently emerged as powerful backbones for recommender systems by reformulating recommendation as a token-level generation task. Despite their promise, we identify a pervasive yet underexplored issue: $\textit{Length Bias}$. Because items are represented by textual descriptions of varying lengths, LLM-based recommenders can be systematically biased in two ways. On the input side, longer item descriptions occupy more tokens in the context and thus receive disproportionately large aggregate attention mass during user preference modeling. On the output side, decoding based on summed autoregressive log-likelihood score inherently disfavors long items. Worse still, conventional length normalization can introduce an additional bias and even degrade recommendation performance. To address this problem, we propose $\textbf{LBR}$ ($\textbf{L}$ength $\textbf{B}$ias $\textbf{R}$eduction), a lightweight and model-agnostic framework for mitigating length bias in LLM-based recommendation. LBR mitigates input-side bias via Length-Aware Attention Calibration, which incorporates a length-dependent offset into attention logits to neutralize attention skew. For the output side, LBR introduces Effective Information Length Normalization, replacing naive token count with an information-theoretic length surrogate derived from the branching structure of the prefix tree. Extensive experiments on three real-world Amazon datasets and two representative LLM-based recommenders demonstrate that LBR substantially alleviates length bias while consistently improving recommendation accuracy and fairness, with negligible additional training and inference overhead (with an average NDCG@5 gain of 16.82%). The code is available at https://github.com/Void-JackLee/LBR.

Figures

Figures reproduced from arXiv: 2607.04270 by the authors.

Figure 1
Figure 1. Illustration of LLM-based recommendations and length bias, wherein item length affects attention accumulation [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Visualization of Trie-Constrained decoding. During [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Illustration of the relationship between an item’s cu [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (6 more)
Figure 5
Figure 5. Figure 5: Illustration of length bias in decoding. Items are partitioned into five groups by token length, with each group [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Visualization of attention weights for the baseline [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Comparison of recommendation distributions [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: Performance comparisons in terms of both recom [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 9
Figure 9. Figure 9: Sensitivity analysis of the hyperparameter [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 10
Figure 10. Figure 10: Impact of item length on average attribution [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

75 extracted references · 26 linked inside Pith

  1. [1]

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Floren- cia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 technical report.arXiv preprint arXiv:2303.08774 (2023)

    arXiv 2023

  2. [2]

    János Aczél, Bruno Forte, and Che Tat Ng. 1974. Why the Shannon and Hartley entropies are ‘natural’.Advances in applied probability6, 1 (1974), 131–146

    1974

  3. [3]

    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

  4. [4]

    Keqin Bao, Jizhi Zhang, Yang Zhang, Xinyue Huo, Chong Chen, and Fuli Feng

  5. [5]

    Decoding matters: Addressing amplification bias and homogeneity issue for llm-based recommendation.arXiv preprint arXiv:2406.14900(2024)

    arXiv 2024

  6. [6]

    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

  7. [7]

    Ethan Bito, Yongli Ren, and Estrid He. 2025. Evaluating Position Bias in Large Language Model Recommendations.arXiv preprint arXiv:2508.02020(2025)

    arXiv 2025

  8. [8]

    Yuwei Cao, Nikhil Mehta, Xinyang Yi, Raghunandan Keshavan, Lukasz Heldt, Lichan Hong, Ed H Chi, and Maheswaran Sathiamoorthy. 2024. Aligning Large Language Models with Recommendation Knowledge.arXiv preprint arXiv:2404.00245(2024)

    arXiv 2024

Show all 75 references
  1. [9]

    Jiawei Chen, Hande Dong, Xiang Wang, Fuli Feng, Meng Wang, and Xiangnan He. 2023. Bias and debias in recommender system: A survey and future directions. ACM Transactions on Information Systems41, 3 (2023), 1–39

    2023

  2. [10]

    Yongjun Chen, Zhiwei Liu, Jia Li, Julian McAuley, and Caiming Xiong. 2022. Intent contrastive learning for sequential recommendation. InProceedings of the ACM web conference 2022. 2172–2182

    2022

  3. [11]

    Yuxin Chen, Junfei Tan, An Zhang, Zhengyi Yang, Leheng Sheng, Enzhi Zhang, Xiang Wang, and Tat-Seng Chua. 2024. On Softmax Direct Preference Optimiza- tion for Recommendation.arXiv preprint arXiv:2406.09215(2024)

    2024 arXiv

  4. [12]

    Yu Cui, Feng Liu, Pengbo Wang, Bohao Wang, Heng Tang, Yi Wan, Jun Wang, and Jiawei Chen. 2024. Distillation Matters: Empowering Sequential Recommenders to Match the Performance of Large Language Models. InProceedings of the 18th ACM Conference on Recommender Systems. 507–517

    2024

  5. [13]

    Sunhao Dai, Chen Xu, Shicheng Xu, Liang Pang, Zhenhua Dong, and Jun Xu

  6. [14]

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

    Bias and unfairness in information retrieval systems: New challenges in the llm era. InProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 6437–6447

  7. [15]

    Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al. 2024. The Llama 3 Herd of Models.arXiv preprint arXiv:2407.21783(2024)

    2024 arXiv

  8. [16]

    Chongming Gao, Ruijun Chen, Shuai Yuan, Kexin Huang, Yuanqing Yu, and Xiangnan He. 2025. Sprec: Self-play to debias llm-based recommendation. In Proceedings of the ACM on Web Conference 2025. 5075–5084

    2025

  9. [17]

    Chongming Gao, Mengyao Gao, Chenxiao Fan, Shuai Yuan, Wentao Shi, and Xiangnan He. 2025. Process-supervised llm recommenders via flow-guided tuning. InProceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1934–1943

    2025

  10. [18]

    Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, et al . 2025. DeepSeek-R1 in- centivizes reasoning in LLMs through reinforcement learning.Nature645, 8081 (2025), 633–638

    2025

  11. [19]

    Balázs Hidasi, Alexandros Karatzoglou, Linas Baltrunas, and Domonkos Tikk

  12. [20]

    Session-based recommendations with recurrent neural networks.arXiv preprint arXiv:1511.06939(2015)

    2015 arXiv

  13. [21]

    Chris Hokamp and Qun Liu. 2017. Lexically constrained decoding for sequence generation using grid beam search.arXiv preprint arXiv:1704.07138(2017)

    2017 arXiv

  14. [22]

    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

    2024

  15. [23]

    Chumeng Jiang, Jiayin Wang, Weizhi Ma, Charles LA Clarke, Shuai Wang, Chuhan Wu, and Min Zhang. 2025. Beyond Utility: Evaluating LLM as Recommender. In Proceedings of the ACM on Web Conference 2025. 3850–3862

    2025

  16. [24]

    Meng Jiang, Keqin Bao, Jizhi Zhang, Wenjie Wang, Zhengyi Yang, Fuli Feng, and Xiangnan He. 2024. Item-side fairness of large language model-based recommen- dation system. InProceedings of the ACM Web Conference 2024. 4717–4726

    2024

  17. [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

  18. [26]

    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 Da...

    2024

  19. [27]

    Dong-Ho Lee, Adam Kraft, Long Jin, Nikhil Mehta, Taibai Xu, Lichan Hong, Ed H Chi, and Xinyang Yi. 2024. STAR: A Simple Training-free Approach for Recommendations using Large Language Models.arXiv preprint arXiv:2410.16458 (2024)

    2024 arXiv

  20. [28]

    Lei Li, Yongfeng Zhang, Dugang Liu, and Li Chen. 2023. Large language models for generative recommendation: A survey and visionary discussions.arXiv preprint arXiv:2309.01157(2023)

    2023 arXiv

  21. [29]

    Xinhang Li, Chong Chen, Xiangyu Zhao, Yong Zhang, and Chunxiao Xing. 2023. E4srec: An elegant effective efficient extensible solution of large language models for sequential recommendation.arXiv preprint arXiv:2312.02443(2023)

    2023 arXiv

  22. [30]

    Bowen Liang, Pidong Wang, and Yuan Cao. 2022. The implicit length bias of label smoothing on beam search decoding.arXiv preprint arXiv:2205.00659(2022)

    2022 arXiv

  23. [31]

    Jiayi Liao, Xiangnan He, Ruobing Xie, Jiancan Wu, Yancheng Yuan, Xingwu Sun, Zhanhui Kang, and Xiang Wang. 2024. RosePO: Aligning LLM-based Recom- menders with Human Values.arXiv preprint arXiv:2410.12519(2024)

    2024 arXiv

  24. [32]

    Jiayi Liao, Sihang Li, Zhengyi Yang, Jiancan Wu, Yancheng Yuan, Xiang Wang, and Xiangnan He. 2024. LLaRA: Large Language-Recommendation Assis- tant. InProceedings of the 47th International ACM SIGIR Conference on Re- search and Development in Information Retrieval(Washington D...

    2024 doi

  25. [33]

    Jianghao Lin, Rong Shan, Chenxu Zhu, Kounianhua Du, Bo Chen, Shigang Quan, Ruiming Tang, Yong Yu, and Weinan Zhang. 2024. Rella: Retrieval-enhanced Preprint, 2026, Hangzhou, China Hongchen Li et al. large language models for lifelong sequential behavior comprehension in recom-...

    2024

  26. [34]

    Siyi Lin, Chongming Gao, Jiawei Chen, Sheng Zhou, Binbin Hu, Yan Feng, Chun Chen, and Can Wang. 2025. How do recommendation models amplify popularity bias? An analysis from the spectral perspective. InProceedings of the Eighteenth ACM International Conference on Web Search and...

    2025

  27. [35]

    Zijie Lin, Yang Zhang, Xiaoyan Zhao, Fengbin Zhu, Fuli Feng, and Tat-Seng Chua. 2025. IGD: Token Decisiveness Modeling via Information Gain in LLMs for Personalized Recommendation.arXiv preprint arXiv:2506.13229(2025)

    2025

  28. [36]

    Sijin Lu, Zhibo Man, Fangyuan Luo, and Jun Wu. 2025. Dual Debiasing in LLM-based Recommendation. InProceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2685–2689

    2025

  29. [37]

    Sichun Luo, Bowei He, Haohan Zhao, Wei Shao, Yanlin Qi, Yinya Huang, Aojun Zhou, Yuxuan Yao, Zongpeng Li, Yuanzhang Xiao, et al. 2025. Recranker: Instruc- tion tuning large language model as ranker for top-k recommendation.ACM Transactions on Information Systems43, 5 (2025), 1–31

    2025

  30. [38]

    Tianhui Ma, Yuan Cheng, Hengshu Zhu, and Hui Xiong. 2023. Large language models are not stable recommender systems.arXiv preprint arXiv:2312.15746 (2023)

    2023 arXiv

  31. [39]

    Vivek Miglani, Aobo Yang, Aram H Markosyan, Diego Garcia-Olano, and Narine Kokhlikyan. 2023. Using captum to explain generative language models.arXiv preprint arXiv:2312.05491(2023)

    2023 arXiv

  32. [40]

    Kenton Murray and David Chiang. 2018. Correcting length bias in neural machine translation.arXiv preprint arXiv:1808.10006(2018)

    2018 arXiv

  33. [41]

    Hyunsoo Na, Minseok Gang, Youngrok Ko, Jinseok Seol, and Sang-goo Lee. 2024. Enhancing Large Language Model Based Sequential Recommender Systems with Pseudo Labels Reconstruction. InFindings of the Association for Computational Linguistics: EMNLP 2024. 7213–7222

    2024

  34. [42]

    Li-Wei Pan, Wei-Ke Pan, Mei-Yan Wei, Hong-Zhi Yin, and Zhong Ming. 2026. A survey on sequential recommendation.Frontiers of Computer Science20, 3 (2026), 2003606

    2026

  35. [43]

    Ivan Provilkov and Andrey Malinin. 2021. Multi-sentence resampling: A simple approach to alleviate dataset length bias and beam-search degradation.arXiv preprint arXiv:2109.06253(2021)

    2021 arXiv

  36. [44]

    Alec Radford, Jeff Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language Models are Unsupervised Multitask Learners. https: //api.semanticscholar.org/CorpusID:160025533

    2019

  37. [45]

    Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. SQuAD: 100,000+ Questions for Machine Comprehension of Text. InProceed- ings of the 2016 Conference on Empirical Methods in Natural Language Processing, Jian Su, Kevin Duh, and Xavier Carreras (Eds.). Ass...

    2016 doi

  38. [46]

    Shashank Rajput, Nikhil Mehta, Anima Singh, Raghunandan Hulikal Keshavan, Trung Vu, Lukasz Heldt, Lichan Hong, Yi Tay, Vinh Tran, Jonah Samost, et al

  39. [47]

    Recommender systems with generative retrieval.Advances in Neural Information Processing Systems36 (2023), 10299–10315

    2023

  40. [48]

    Shiori Sagawa, Pang Wei Koh, Tatsunori B Hashimoto, and Percy Liang. 2019. Distributionally robust neural networks for group shifts: On the importance of regularization for worst-case generalization.arXiv preprint arXiv:1911.08731 (2019)

    2019 arXiv

  41. [49]

    Felix Stahlberg. 2020. Neural machine translation: A review.Journal of Artificial Intelligence Research69 (2020), 343–418

    2020

  42. [50]

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

  43. [51]

    InProceedings of the 28th ACM international conference on information and knowledge management

    BERT4Rec: Sequential recommendation with bidirectional encoder rep- resentations from transformer. InProceedings of the 28th ACM international conference on information and knowledge management. 1441–1450

  44. [52]

    Zhongxiang Sun, Zihua Si, Xiaoxue Zang, Kai Zheng, Yang Song, Xiao Zhang, and Jun Xu. 2024. Large Language Models Enhanced Collaborative Filtering. arXiv preprint arXiv:2403.17688(2024)

    2024 arXiv

  45. [53]

    Juntao Tan, Shuyuan Xu, Wenyue Hua, Yingqiang Ge, Zelong Li, and Yongfeng Zhang. 2024. Idgenrec: Llm-recsys alignment with textual id learning. InProceed- ings of the 47th international ACM SIGIR conference on research and development in information retrieval. 355–364

    2024

  46. [54]

    Jiaxi Tang and Ke Wang. 2018. Personalized top-n sequential recommenda- tion via convolutional sequence embedding. InProceedings of the eleventh ACM international conference on web search and data mining. 565–573

    2018

  47. [55]

    Qwen Team et al. 2024. Qwen2 technical report.arXiv preprint arXiv:2407.10671 2, 3 (2024)

    2024 arXiv

  48. [56]

    Ashish Vaswani. 2017. Attention is all you need.arXiv preprint arXiv:1706.03762 (2017)

    2017 arXiv

  49. [57]

    Bohao Wang, Jiawei Chen, Changdong Li, Sheng Zhou, Qihao Shi, Yang Gao, Yan Feng, Chun Chen, and Can Wang. 2024. Distributionally Robust Graph-based Recommendation System. InProceedings of the ACM on Web Conference 2024. 3777–3788

    2024

  50. [58]

    Bohao Wang, Jiawei Chen, Feng Liu, Changwang Zhang, Jun Wang, Canghong Jin, Chun Chen, and Can Wang. 2026. Does LLM Focus on the Right Words? Mitigating Context Bias in LLM-based Recommenders. arXiv:2510.10978 [cs.IR] https://arxiv.org/abs/2510.10978

    2026

  51. [59]

    Bohao Wang, Feng Liu, Jiawei Chen, Xingyu Lou, Changwang Zhang, Jun Wang, Yuegang Sun, Yan Feng, Chun Chen, and Can Wang. 2025. Msl: Not all tokens are what you need for tuning llm as a recommender. InProceedings of the 48th International ACM SIGIR Conference on Research and D...

    2025

  52. [60]

    Bohao Wang, Feng Liu, Changwang Zhang, Jiawei Chen, Yudi Wu, Sheng Zhou, Xingyu Lou, Jun Wang, Yan Feng, Chun Chen, et al. 2025. Llm4dsr: Leveraging large language model for denoising sequential recommendation.ACM Transac- tions on Information Systems44, 1 (2025), 1–32

    2025

  53. [61]

    Hangyu Wang, Jianghao Lin, Xiangyang Li, Bo Chen, Chenxu Zhu, Ruiming Tang, Weinan Zhang, and Yong Yu. 2024. Flip: Fine-grained alignment between id-based models and pretrained language models for ctr prediction. InProceedings of the 18th ACM conference on recommender systems. 94–104

    2024

  54. [62]

    Lei Wang and Ee-Peng Lim. 2023. Zero-shot next-item recommendation using large pretrained language models.arXiv preprint arXiv:2304.03153(2023)

    2023 arXiv

  55. [63]

    Likang Wu, Zhi Zheng, Zhaopeng Qiu, Hao Wang, Hongchao Gu, Tingjia Shen, Chuan Qin, Chen Zhu, Hengshu Zhu, Qi Liu, et al . 2024. A survey on large language models for recommendation.World Wide Web27, 5 (2024), 60

    2024

  56. [64]

    Yonghui Wu, Mike Schuster, Zhifeng Chen, Quoc V Le, Mohammad Norouzi, Wolfgang Macherey, Maxim Krikun, Yuan Cao, Qin Gao, Klaus Macherey, et al

  57. [65]

    Google’s neural machine translation system: Bridging the gap between human and machine translation.arXiv preprint arXiv:1609.08144(2016)

    2016 arXiv

  58. [66]

    Xu Xie, Fei Sun, Zhaoyang Liu, Shiwen Wu, Jinyang Gao, Jiandong Zhang, Bolin Ding, and Bin Cui. 2022. Contrastive learning for sequential recommendation. In 2022 IEEE 38th international conference on data engineering (ICDE). IEEE, 1259– 1273

    2022

  59. [67]

    Shi Xuewen, Huang Heyan, Jian Ping, and Tang Yi-Kun. 2021. Reducing length bias in scoring neural machine translation via a causal inference method. In Proceedings of the 20th Chinese National Conference on Computational Linguistics. 874–885

    2021

  60. [68]

    An Yang, Beichen Zhang, Binyuan Hui, Bofei Gao, Bowen Yu, Chengpeng Li, Dayiheng Liu, Jianhong Tu, Jingren Zhou, Junyang Lin, Keming Lu, Mingfeng Xue, Runji Lin, Tianyu Liu, Xingzhang Ren, and Zhenru Zhang. 2024. Qwen2.5-Math Technical Report: Toward Mathematical Expert Model ...

    2024 arXiv

  61. [69]

    Yilin Yang, Liang Huang, and Mingbo Ma. 2018. Breaking the beam search curse: A study of (re-) scoring methods and stopping criteria for neural machine translation.arXiv preprint arXiv:1808.09582(2018)

    2018 arXiv

  62. [70]

    Zhengyi Yang, Xiangnan He, Jizhi Zhang, Jiancan Wu, Xin Xin, Jiawei Chen, and Xiang Wang. 2023. A generic learning framework for sequential recommenda- tion with distribution shifts. InProceedings of the 46th International ACM SIGIR Conference on Research and Development in In...

    2023

  63. [71]

    Junjie Zhang, Yupeng Hou, Ruobing Xie, Wenqi Sun, Julian McAuley, Wayne Xin Zhao, Leyu Lin, and Ji-Rong Wen. 2024. Agentcf: Collaborative learning with autonomous language agents for recommender systems. InProceedings of the ACM Web Conference 2024. 3679–3689

    2024

  64. [72]

    Yang Zhang, Juntao You, Yimeng Bai, Jizhi Zhang, Keqin Bao, Wenjie Wang, and Tat-Seng Chua. 2024. Causality-enhanced behavior sequence modeling in LLMs for personalized recommendation.arXiv preprint arXiv:2410.22809(2024)

    2024 arXiv

  65. [73]

    Bowen Zheng, Yupeng Hou, Hongyu Lu, Yu Chen, Wayne Xin Zhao, Ming Chen, and Ji-Rong Wen. 2024. Adapting large language models by integrating collaborative semantics for recommendation. In2024 IEEE 40th International Conference on Data Engineering (ICDE). IEEE, 1435–1448

    2024

  66. [74]

    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

  67. [75]

    Yaochen Zhu, Liang Wu, Qi Guo, Liangjie Hong, and Jundong Li. 2024. Collab- orative large language model for recommender systems. InProceedings of the ACM on Web Conference 2024. 3162–3172. A Hyperparameter Study on𝛼in Length Normalization In this section, we present experimen...

    2024

Pith tools

Reviewed July 11, 2026 · model on record in the stance chip above.