Pith. sign in

REVIEW 4 major objections 5 minor 72 references

Unsupervised dense retrieval with conterfactual contrastive learning

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Counterfactual training shields dense retrievers from ranking attacks.

desk verdict Useful unsupervised key-passage extraction via Shapley values, but the adversarial robustness headline rests on an underspecified construction that may just be random noise. read the letter →

arxiv 2412.20756 v1 pith:ZLI2ZI6V submitted 2024-12-30 cs.IR

Haitian Chen , Qingyao Ai , Xiao Wang , Yiqun Liu , Fen Lin , Qin Liu This is my paper · ORCID
classification cs.IR
keywords denseretrievalcounterfactuallearningShapleyvalueadversarialrobustnesskeypassageextractioncontrastiveMSMARCO
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

Dense retrieval models rank documents by embedding similarity, but they are easy to fool: small synonym substitutions can push an irrelevant document to the top. This paper argues that the root problem is insensitivity to fine-grained relevance, and proposes to fix it by forcing the model to react strongly when the passage that actually answers a query is altered, while staying calm when irrelevant passages change. To find that key passage without any passage-level labels, the authors treat a document's passages as players in a coalitional game and rank them by Shapley value computed from the retriever's own relevance scores. The same counterfactual passages then become training pivots in a contrastive loss, so the retriever learns to distinguish a true positive from a slightly damaged version and from a negative. The result, on MS MARCO document retrieval, is a model that locates key passages better than simpler deletion-based baselines and loses less ranking accuracy under four adversarial attacks than standard adversarial training and a certified defense.

What carries the argument

The load-bearing object is the Shapley value of a passage in a coalitional game whose players are the document's passages and whose value function $v(P)$ is the retriever's relevance score on the document made of the passage subset $P$. Equation (4) computes each passage's contribution as the weighted average of marginal score changes over all subsets, and the paper approximates the exponential sum by computing over non-overlapping windows and merging the results with a moving average. The companion mechanism is a stack of counterfactual contrastive losses: partial counterfactuals (one sentence removed from the key passage), full counterfactuals (key passage removed), and adversarial counterfactuals (5% of words replaced) serve as graded pivots between the positive and negative documents, forcing the embedding space to respect $f(q,d^+) > f(q,d') > f(q,d^*) > f(q,d^-)$ for counterfactuals and $f(q,d^+) > f(q,d_{adv}) > f(q,d^-)$ for attacks. Shapley values also set the loss weights, and a coupling-learning strategy that adapts the weights during training gives the best key-passage extraction and robustness.

What would settle it

Two concrete tests would settle the claim. First, recompute the Shapley passage ranking after shuffling passage order and check whether the top-ranked passage still matches the labeled relevant passage. Second, regenerate adversarial counterfactuals with the argmax in Eq. (8) rather than random 5% replacement and rerun the four attacks: if the robustness gap over adversarial training shrinks or vanishes, the defense rested on easy counterfactuals.

Watch

Extended reading notes

Core claim

The paper's central claim is that the vulnerability of dense retrievers and their opacity can be addressed by making the relevance score obey an explicit sensitivity profile: high variance when the key passage of a relevant document is removed or corrupted, low variance when irrelevant passages are touched. The authors show that Shapley values computed over passages with the retriever's score as the value function identify the key passage more accurately than delta-rank or delta-relevance baselines, and that this signal can drive training even when passage labels are simulated by an LLM. They then use partial, full, and adversarially constructed counterfactual documents as graded pivots in three contrastive losses, so the regularized model places the positive above partial counterfactuals, full counterfactuals, adversarial counterfactuals, and negatives in that order. On term spamming, PRADA, PAT, and MCARA, the regularized DPR loses only 10.1 to 16.1 percent of MRR@10d, a smaller drop than adversarial training (13.5 to 19.2 percent) and the certified defense CertDR (12.4 to 17.3 percent), and it beats the attack-specific PIAT on two of the four attacks. The same regularization does not degrade retrieval effectiveness, which is the practical payoff: robustness and interpretability come from the same training signal.

Load-bearing premise

The method assumes that a retriever's relevance score on arbitrary subsets of a document's passages is a faithful measure of what those passages contribute, so deleting a passage lowers the score exactly when that passage matters.

Editorial extensions

If this is right

  • Key passage extraction becomes a byproduct of retrieval: any dense retriever can identify the passage that justifies a query without passage-level relevance labels.
  • The robustness gain does not require training on attack-specific adversarial examples, so the defense generalizes to attack methods the model never saw, whereas PIAT needs attack-generated training data.
  • The regularization leaves standard retrieval performance intact and even slightly improves MRR@10d on MS MARCO-doc and NDCG@10 on TREC DL2019, so robustness is not bought at the cost of effectiveness.
  • Shapley-based loss weighting beats relevance-score weighting, so the quality of the passage attribution directly controls the size of the robustness improvement.
  • Ablations show that the combination of pseudo-positive and hard-negative counterfactual losses drives key-passage extraction, while the adversarial counterfactual loss drives robustness, and only the full loss stack achieves the best of both.

Reading between the lines

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

  • Beyond the paper: the Shapley attribution is only as sound as the score function it is computed from, so a natural next check is whether the same passage ranking survives different document segmentations and window sizes; the moving-average merge used here partly masks that dependency.
  • Beyond the paper: because the defense does not require attack-specific training examples, the same counterfactual regularization could be inserted into late-interaction or generative retrieval models as a label-free robustness layer.
  • Beyond the paper: the coupling-learning weighting suggests that alpha and beta can be learned rather than tuned, pointing toward a curriculum that gradually moves from partial to full to adversarial counterfactuals.
  • Beyond the paper: since the method produces a causal, counterfactual rationale for each ranking, comparing those rationales against multi-passage human annotations on a dataset richer than a single relevant passage per document would be a sharper test of the explainability claim.
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

4 major / 5 minor

Summary. The paper proposes a counterfactual contrastive learning method for dense retrieval. The authors use Shapley values over document passages to identify key passages without passage-level relevance annotations, then construct three types of counterfactual documents (partial, full, and adversarial) and add contrastive losses that push positive documents above counterfactuals above negatives. Experiments on MS MARCO document and passage datasets report improved key-passage extraction MRR@10p for DPR and ANCE, no degradation in document retrieval, and smaller relative performance drops under four adversarial attack methods, leading the authors to claim state-of-the-art robustness.

Significance. If the mechanism were fully specified and reproducible, the paper would make a useful empirical contribution: it addresses both explainability (key passage extraction without passage labels) and robustness of dense retrievers in one framework, and it evaluates across six retrieval models and four attack methods. The ablations in Tables 5 and 8 are informative, and the comparison against AT, CertDR, and PIAT is a reasonable first step. However, the central mechanisms are under-specified, and one of the headline claims in the abstract is stronger than Table 7 supports.

major comments (4)
  1. [Section 3.2, Eq. (4)] The Shapley-value key passage extraction is not reproducible as written. The manuscript does not specify how the relevance function v(P) is evaluated for an arbitrary subset of passages P when the retriever scores whole documents, how the exponential number of coalitions is sampled or approximated, or how the 'Non-overlap' and 'Merge' procedures described in Section 4.1 relate to Eq. (4). Since Table 2's central claim that Shapley values outperform δrank and δrel depends on this computation, please provide a precise algorithm, including the subset-to-document mapping, the approximation strategy, and the number of samples used, and ideally release the code.
  2. [Section 3.3, Eqs. (8)-(9) and Section 4.3] The adversarial counterfactual document d_adv is defined as the argmax of f(q,d') - f(q,d) over B(d,epsilon), but the only implementation detail given is that epsilon is 5% and B(d,epsilon) is described as a set where words are 'randomly replaced.' Random replacement is not an argmax search and can easily produce documents that violate the ordering f(q,d+) > f(q,d_adv) > f(q,d-) asserted in Eq. (9). Because Table 8 attributes the largest robustness gains to the L_adv term, the reported robustness improvements may be driven by exposure to generic noisy documents rather than by the claimed adversarial counterfactuals. Please specify the exact d_adv construction procedure, verify the Eq. (9) ordering on training data, and report the empirical fraction of examples for which it holds.
  3. [Abstract and Table 7] The abstract and conclusion claim that the regularized dense retrieval models 'surpass the state-of-the-art anti-attack methods,' but Table 7 shows that DPR_counter outperforms PIAT only on TS and PRADA, while PIAT is significantly better under PAT and MCARA. The claim should be weakened to a statement such as 'comparable to or better than PIAT on some attacks without requiring attack-specific training data,' or additional evidence should be supplied to justify the stronger wording.
  4. [Table 7] The main robustness comparison against AT, CertDR, and PIAT is reported only for DPR_counter; ANCE_counter appears in Table 6 but is not included in Table 7. Since the abstract refers to 'regularized dense retrieval models' in the plural, including ANCE_counter in the Table 7 comparison would substantially strengthen the generality of the robustness claim.
minor comments (5)
  1. [Title and abstract] The title contains a typo, 'conterfactual,' and the abstract contains 'fine-graned'; both should be corrected.
  2. [Section 3.3, Eq. (10)] The sentence introducing L_neg says the loss maximizes the similarity between (q,d) and (q,d'), but the formula uses (q,d+) as the anchor; the text should be aligned with the equation.
  3. [Section 3.3, Eq. (13)] The text says the final loss is a weighted sum of 'all three loss functions,' but the expression contains L_cla, L_neg, L_pos, and L_adv; please clarify the grouping or change the wording.
  4. [Section 3.3, loss weight strategies] The 'couple learning' strategy is attributed to reference [33], whose title is 'Perturbation-Invariant Adversarial Training for Neural Ranking Models'; this citation does not appear to describe a couple-learning method, so please verify and, if needed, cite the correct source.
  5. [Title and Section 1] The term 'unsupervised' should be qualified: the method does not use passage-level relevance labels, but it does use query-document relevance labels and hard negatives, so it is more accurately described as 'without passage-level relevance annotations.'

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the Shapley-based pseudo-labeling loop is a self-training bootstrap, but the headline claims are validated on independent human passage labels and external attack methods.

full rationale

The paper's derivation chain is not circular in the sense of a prediction reducing to its inputs by construction. The key-passage extraction method uses Shapley values computed from the model's own relevance scores f(q,d), and the counterfactual contrastive losses (Eqs. 10-12) then train the model to be sensitive to the passages selected by that same procedure. This is a self-referential training-signal design: the model's own scores supply pseudo-labels for which passages are important. However, the loop is not closed. The key-passage claim is evaluated against human-annotated MS MARCO relevant passages via MRR@10p, not against the model's own Shapley ranking; the robustness claim is tested under four external attack methods (TS, PRADA, PAT, MCARA) and compared with external baselines AT, CertDR, and PIAT. The inequalities in Eqs. (6) and (9) are training targets, not evaluation metrics, and the hyperparameters alpha and beta are chosen from a small fixed set of strategies rather than fitted to the test metric. The gap between the argmax definition of d_adv in Eq. (8) and the reported 'randomly replaced' 5% implementation is a verification and correctness concern, but it does not make the robustness evaluation circular, because the test attacks are generated by independent procedures. No load-bearing self-citation or imported uniqueness theorem appears. Accordingly, the appropriate finding is no significant circularity.

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

The method relies on several hand-chosen hyperparameters and domain assumptions about the reliability of Shapley values and monotone relevance scores. No new physical or conceptual entities are introduced; the 'counterfactual passages' are constructed artifacts, not independent postulates.

free parameters (5)
  • loss weight hyperparameters alpha, beta = three strategies: relevance score (alpha=1-r, beta=r), Shapley value (alpha=1-s', beta=s'), coupling learning [33]
    Chosen by comparing three strategies on the validation set; the coupling learning variant is reported as best.
  • segmentation window size = 128 (vs 64)
    Selected after Table 1 showed 128 preserves 95.8% of positive passages and yields best MRR@10p.
  • modification type for counterfactual documents = deletion (vs modification, replacement)
    Selected after Table 1 showed deletion gives the largest relevance drop and best extraction performance.
  • adversarial perturbation ratio epsilon = 5%
    Set by hand; no sensitivity analysis is provided.
  • number of hard negatives per query = 7
    Standard choice; no ablation.
assumptions (4)
  • domain assumption The dense retriever's relevance score on arbitrary subsets of passages (v(P)) defines a valid coalitional game for Shapley values.
    Invoked for Eq. (4); the paper does not specify how v(P) is computed for a subset of passages.
  • domain assumption Deleting the key passage reduces relevance more than deleting other passages, and this ordering is consistent across the model.
    Underpins the inequalities in Eq. (6) and the design of L_neg.
  • domain assumption Adversarial counterfactual documents constructed by argmax over B(d, epsilon) satisfy f(q,d+) > f(q,d_adv) > f(q,d*).
    Assumed in Eq. (9) and L_adv, but the implementation appears to use random replacement, so the inequality may not hold.
  • standard math The contrastive loss in Eq. (3) with in-batch negatives is a valid training objective for dense retrieval.
    Standard InfoNCE-style loss for retrieval.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Unsupervised dense retrieval with conterfactual contrastive learning." pith.science (2026). https://pith.science/paper/ZLI2ZI6V

@misc{pith2026241220756,
  author       = {Pith},
  title        = {Pith review of: Unsupervised dense retrieval with conterfactual contrastive learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZLI2ZI6V}},
  note         = {Machine review of arXiv:2412.20756}
}
read the original abstract

Efficiently retrieving a concise set of candidates from a large document corpus remains a pivotal challenge in Information Retrieval (IR). Neural retrieval models, particularly dense retrieval models built with transformers and pretrained language models, have been popular due to their superior performance. However, criticisms have also been raised on their lack of explainability and vulnerability to adversarial attacks. In response to these challenges, we propose to improve the robustness of dense retrieval models by enhancing their sensitivity of fine-graned relevance signals. A model achieving sensitivity in this context should exhibit high variances when documents' key passages determining their relevance to queries have been modified, while maintaining low variances for other changes in irrelevant passages. This sensitivity allows a dense retrieval model to produce robust results with respect to attacks that try to promote documents without actually increasing their relevance. It also makes it possible to analyze which part of a document is actually relevant to a query, and thus improve the explainability of the retrieval model. Motivated by causality and counterfactual analysis, we propose a series of counterfactual regularization methods based on game theory and unsupervised learning with counterfactual passages. Experiments show that, our method can extract key passages without reliance on the passage-level relevance annotations. Moreover, the regularized dense retrieval models exhibit heightened robustness against adversarial attacks, surpassing the state-of-the-art anti-attack methods.

Figures

Figures reproduced from arXiv: 2412.20756 by the authors.

Figure 1
Figure 1. Illustration of the counterfactual and adversarial doc [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 1
Figure 1. Partial/Full Counterfactual Document. Formally, for a query￾document pair (𝑞, 𝑑+ ), we define its key passage as 𝑝 + , and a set of N negative documents as 𝐷 −. The idea of contrastive learning to train the dense retrieval models to create query and document representations such that the following inequality between positive document 𝑑 + and negative document 𝑑 − ∈ 𝐷 − holds: 𝑓 (𝑞, 𝑑+ ) > 𝑓 (𝑞, 𝑑− ). (5) We define p… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

72 extracted references · 33 canonical work pages

  1. [1]

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Flo- rencia 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]

    2008.Models in cooperative game theory

    Rodica Branzei, Dinko Dimitrov, and Stef Tijs. 2008.Models in cooperative game theory. V ol. 556. Springer Science & Business Media

    work page 2008

  3. [3]

    Xuanang Chen, Ben He, Zheng Ye, Le Sun, and Yingfei Sun. 2023. Towards Imperceptible Document Manipulations against Neural Ranking Models. arXiv preprint arXiv:2305.01860 (2023)

    arXiv 2023

  4. [4]

    Seungtaek Choi, Haeju Park, Jinyoung Yeo, and Seung-won Hwang. 2020. Less is More: Attention Supervision with Counterfactuals for Text Classification. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020, Online, November 16-20, 2020, Bonnie Webber, Trevor Cohn, Yulan He, and Yang Liu (Eds.). Association fo...

    work page doi:10.18653/v1/2020.emnlp-main.543 2020

  5. [5]

    Jiequan Cui, Shu Liu, Liwei Wang, and Jiaya Jia. 2021. Learnable boundary guided adversarial training. In Proceedings of the IEEE/CVF international conference on computer vision. 15721–15730

    work page 2021

  6. [6]

    Zhuyun Dai and Jamie Callan. 2020. Context-aware term weighting for first stage passage retrieval. In Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval. 1533–1536

    work page 2020

  7. [7]

    Zhuyun Dai, Chenyan Xiong, Jamie Callan, and Zhiyuan Liu. 2018. Convolu- tional Neural Networks for Soft-Matching N-Grams in Ad-hoc Search. In Pro- ceedings of the Eleventh ACM International Conference on Web Search and Data Mining, WSDM 2018, Marina Del Rey, CA, USA, February 5-9, 2018 , Yi Chang, Chengxiang Zhai, Yan Liu, and Yoelle Maarek (Eds.). ACM, ...

    arXiv 2018

  8. [8]

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Associa- tion for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, MN, USA, June 2-7, 2019, ...

    2019

Show all 72 references
  1. [9]

    Luyu Gao, Zhuyun Dai, Zhen Fan, and Jamie Callan. 2020. Complementing Lexical Retrieval with Semantic Residual Embedding. CoRR abs/2004.13969 (2020). arXiv:2004.13969 https://arxiv.org/abs/2004.13969

    2020 arXiv

  2. [10]

    Ian J Goodfellow, Jonathon Shlens, and Christian Szegedy. 2014. Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572 (2014)

    2014 arXiv

  3. [11]

    Yash Goyal, Ziyan Wu, Jan Ernst, Dhruv Batra, Devi Parikh, and Stefan Lee

  4. [12]

    Bruce Croft

    Jiafeng Guo, Yixing Fan, Qingyao Ai, and W. Bruce Croft. 2016. A Deep Rel- evance Matching Model for Ad-hoc Retrieval. In Proceedings of the 25th ACM International Conference on Information and Knowledge Management, CIKM 2016, Indianapolis, IN, USA, October 24-28, 2016 , Sneha...

    2016

  5. [13]

    Zoltán Gyöngyi, Hector Garcia-Molina, et al. 2005. Web Spam Taxonomy.. In AIRWeb, V ol. 5. 39–47

    2005

  6. [14]

    Kai Hui, Andrew Yates, Klaus Berberich, and Gerard de Melo. 2017. PACRR: A Position-Aware Neural IR Model for Relevance Matching. InProceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, EMNLP 2017, Copenhagen, Denmark, September 9-11, 2017 , M...

    2017 doi

  7. [15]

    Robin Jia, Aditi Raghunathan, Kerem Göksel, and Percy Liang. 2019. Certified robustness to adversarial word substitutions. arXiv preprint arXiv:1909.00986 (2019)

    2019 arXiv

  8. [16]

    Johansson, Uri Shalit, and David A

    Fredrik D. Johansson, Uri Shalit, and David A. Sontag. 2016. Learning Rep- resentations for Counterfactual Inference. In Proceedings of the 33nd Inter- national Conference on Machine Learning, ICML 2016, New York City, NY, USA, June 19-24, 2016 (JMLR Workshop and Conference Pr...

    2016

  9. [17]

    Vladimir Karpukhin, Barlas O˘guz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih. 2020. Dense passage retrieval for open- domain question answering. arXiv preprint arXiv:2004.04906 (2020)

    2020 arXiv

  10. [18]

    Vladimir Karpukhin, Barlas Oguz, Sewon Min, Patrick S. H. Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih. 2020. Dense Passage Retrieval for Open-Domain Question Answering. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, ...

    2020

  11. [19]

    Omar Khattab and Matei Zaharia. 2020. ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT. InProceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval, SIGIR 2020, Virtual Event, Ch...

    2020

  12. [20]

    Kusner, Joshua R

    Matt J. Kusner, Joshua R. Loftus, Chris Russell, and Ricardo Silva. 2017. Coun- terfactual Fairness. In Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, Decem- ber 4-9, 2017, Long Beach, CA, USA , Isabelle G...

    2017

  13. [21]

    Kenton Lee, Ming-Wei Chang, and Kristina Toutanova. 2019. Latent Retrieval for Weakly Supervised Open Domain Question Answering. InProceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28- August 2, 2019, Volume 1:...

    2019 doi

  14. [22]

    Mike Lewis, Marjan Ghazvininejad, Gargi Ghosh, Armen Aghajanyan, Sida Wang, and Luke Zettlemoyer. 2020. Pre-training via Paraphrasing. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, Decemb...

    2020

  15. [23]

    Patrick S. H. Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2020. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. In Advanc...

    2020

  16. [24]

    Ke Liang, Yue Liu, Sihang Zhou, Wenxuan Tu, Yi Wen, Xihong Yang, Xiangjun Dong, and Xinwang Liu. 2024. Knowledge Graph Contrastive Learning Based on Relation-Symmetrical Structure. IEEE Trans. Knowl. Data Eng. 36, 1 (2024), 226–238. https://doi.org/10.1109/TKDE.2023.3282989

    2024

  17. [25]

    Ke Liang, Lingyuan Meng, Meng Liu, Yue Liu, Wenxuan Tu, Siwei Wang, Sihang Zhou, and Xinwang Liu. 2023. Learn from Relational Correlations and Periodic Events for Temporal Knowledge Graph Reasoning. In Proceedings of the 46th In- ternational ACM SIGIR Conference on Research an...

    2023

  18. [26]

    Zujie Liang, Weitao Jiang, Haifeng Hu, and Jiaying Zhu. 2020. Learning to Contrast the Counterfactual Samples for Robust Visual Question Answering. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020, Online, November 16-20, 20...

    2020 doi

  19. [27]

    Jiawei Liu, Yangyang Kang, Di Tang, Kaisong Song, Changlong Sun, Xiaofeng Wang, Wei Lu, and Xiaozhong Liu. 2022. Order-Disorder: Imitation Adversarial Attacks for Black-box Neural Ranking Models. In Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications S...

    2022

  20. [28]

    Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. RoBERTa: A Robustly Optimized BERT Pretraining Approach. CoRR abs/1907.11692 (2019). arXiv:1907.11692 http://arxiv.org/abs/1907.11692

    2019 arXiv

  21. [29]

    Yu-An Liu, Ruqing Zhang, Jiafeng Guo, Wei Chen, and Xueqi Cheng. 2023. On the Robustness of Generative Retrieval Models: An Out-of-Distribution Perspective. CoRR abs/2306.12756 (2023). https://doi.org/10.48550/ARXIV .2306.12756 arXiv:2306.12756

    2023 doi arXiv

  22. [30]

    Yu-An Liu, Ruqing Zhang, Jiafeng Guo, Maarten de Rijke, Wei Chen, Yixing Fan, and Xueqi Cheng. 2023. Black-box Adversarial Attacks against Dense Retrieval Models: A Multi-view Contrastive Learning Method. In Proceed- ings of the 32nd ACM International Conference on Information...

    2023

  23. [31]

    Yu-An Liu, Ruqing Zhang, Jiafeng Guo, Maarten de Rijke, Wei Chen, Yixing Fan, and Xueqi Cheng. 2023. Black-box Adversarial Attacks against Dense Retrieval Models: A Multi-view Contrastive Learning Method. In Proceedings of the 32nd ACM International Conference on Information a...

    2023

  24. [32]

    Yu-An Liu, Ruqing Zhang, Jiafeng Guo, Maarten de Rijke, Wei Chen, Yixing Fan, and Xueqi Cheng. 2023. Topic-oriented Adversarial Attacks against Black-box Neural Ranking Models. arXiv preprint arXiv:2304.14867 (2023)

    2023 arXiv

  25. [33]

    Yu-An Liu, Ruqing Zhang, Mingkun Zhang, Wei Chen, Maarten de Rijke, Ji- afeng Guo, and Xueqi Cheng. 2023. Perturbation-Invariant Adversarial Training for Neural Ranking Models: Improving the Effectiveness-Robustness Trade-Off. arXiv preprint arXiv:2312.10329 (2023)

    2023 arXiv

  26. [34]

    Yi Luan, Jacob Eisenstein, Kristina Toutanova, and Michael Collins. 2021. Sparse, Dense, and Attentional Representations for Text Retrieval. Trans. Assoc. Comput. Linguistics 9 (2021), 329–345. https://doi.org/10.1162/TACL_A_00369

    2021 doi

  27. [35]

    Yi Luan, Jacob Eisenstein, Kristina Toutanova, and Michael Collins. 2021. Sparse, dense, and attentional representations for text retrieval. Transactions of the Asso- ciation for Computational Linguistics 9 (2021), 329–345

    2021

  28. [36]

    Scott M Lundberg and Su-In Lee. 2017. A unified approach to interpreting model predictions. Advances in neural information processing systems 30 (2017)

    2017

  29. [37]

    Sean MacAvaney, Andrew Yates, Arman Cohan, and Nazli Goharian. 2019. CEDR: Contextualized Embeddings for Document Ranking. In Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Informa- tion Retrieval, SIGIR 2019, Paris, France, July 21-2...

    2019

  30. [38]

    Aleksander Madry, Aleksandar Makelov, Ludwig Schmidt, Dimitris Tsipras, and Adrian Vladu. 2017. Towards deep learning models resistant to adversarial attacks. arXiv preprint arXiv:1706.06083 (2017)

    2017 arXiv

  31. [39]

    Bhaskar Mitra and Nick Craswell. 2018. An Introduction to Neural Information Retrieval. Found. Trends Inf. Retr.13, 1 (2018), 1–126. https://doi.org/10.1561/ 1500000061

    2018

  32. [40]

    Tri Nguyen, Mir Rosenberg, Xia Song, Jianfeng Gao, Saurabh Tiwary, Rangan Majumder, and Li Deng. 2016. MS MARCO: A human generated machine reading comprehension dataset. choice 2640 (2016), 660

    2016

  33. [41]

    Yulei Niu, Kaihua Tang, Hanwang Zhang, Zhiwu Lu, Xian-Sheng Hua, and Ji- Rong Wen. 2021. Counterfactual VQA: A Cause-Effect Look at Language Bias. In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, virtual, June 19-25, 2021. Computer Vision Foundation / ...

    2021

  34. [42]

    Rodrigo Nogueira, Jimmy Lin, and AI Epistemic. 2019. From doc2query to docTTTTTquery. Online preprint 6 (2019), 2

    2019

  35. [43]

    Rodrigo Frassetto Nogueira and Kyunghyun Cho. 2019. Passage Re-ranking with BERT. CoRR abs/1901.04085 (2019). arXiv:1901.04085 http://arxiv.org/abs/ 1901.04085

    2019 arXiv

  36. [44]

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

    2019

  37. [45]

    Prafull Prakash, Julian Killingback, and Hamed Zamani. 2021. Learning Robust Dense Retrieval Models from Incomplete Relevance Labels. In SIGIR ’21: The 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, Virtual Event, Canada, July 11-...

    2021

  38. [46]

    Bruce Croft, and Mohit Iyyer

    Chen Qu, Liu Yang, Cen Chen, Minghui Qiu, W. Bruce Croft, and Mohit Iyyer

  39. [47]

    Yingqi Qu, Yuchen Ding, Jing Liu, Kai Liu, Ruiyang Ren, Wayne Xin Zhao, Daxiang Dong, Hua Wu, and Haifeng Wang. 2020. RocketQA: An optimized training approach to dense passage retrieval for open-domain question answering. arXiv preprint arXiv:2010.08191 (2020)

    2020 arXiv

  40. [48]

    Aditi Raghunathan, Jacob Steinhardt, and Percy Liang. 2018. Certified defenses against adversarial examples. arXiv preprint arXiv:1801.09344 (2018)

    2018 arXiv

  41. [49]

    Juan Ramos et al. 2003. Using tf-idf to determine word relevance in document queries. In Proceedings of the first instructional conference on machine learning, V ol. 242. New Jersey, USA, 29–48

    2003

  42. [50]

    Why should i trust you?

    Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. " Why should i trust you?" Explaining the predictions of any classifier. InProceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. 1135–1144

    2016

  43. [51]

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

    2009

  44. [52]

    Avanti Shrikumar, Peyton Greenside, and Anshul Kundaje. 2017. Learning impor- tant features through propagating activation differences. In International confer- ence on machine learning. PMLR, 3145–3153

    2017

  45. [53]

    Yongho Song, Dahyun Lee, Kyungjae Lee, and Jinyeong Yeo. 2023. Re- visiting Dense Retrieval with Unanswerable Counterfactuals. arXiv preprint arXiv:2304.03031 (2023)

    2023 arXiv

  46. [54]

    Goodfellow, and Rob Fergus

    Christian Szegedy, Wojciech Zaremba, Ilya Sutskever, Joan Bruna, Dumitru Erhan, Ian J. Goodfellow, and Rob Fergus. 2014. Intriguing properties of neural networks. In 2nd International Conference on Learning Representations, ICLR 2014, Banff, AB, Canada, April 14-16, 2014, Conf...

    2014 arXiv

  47. [55]

    Bing Tian, Yixin Cao, Yong Zhang, and Chunxiao Xing. 2022. Debiasing NLU Models via Causal Intervention and Counterfactual Reasoning. In Thirty-Sixth AAAI Conference on Artificial Intelligence, AAAI 2022, Thirty-Fourth Conference on Innovative Applications of Artificial Intell...

    2022 doi

  48. [56]

    Shaokun Wang, Tian Gan, Yuan Liu, Jianlong Wu, Yuan Cheng, and Liqiang Nie. 2023. Micro-Influencer Recommendation by Multi-Perspective Account Representation Learning. IEEE Trans. Multim. 25 (2023), 2749–2760. https: //doi.org/10.1109/TMM.2022.3151029

    2023

  49. [57]

    Shaokun Wang, Tian Gan, Yuan Liu, Li Zhang, Jianlong Wu, and Liqiang Nie. 2022. Discover Micro-Influencers for Brands via Better Understanding. IEEE Trans. Multim. 24 (2022), 2595–2605. https://doi.org/10.1109/TMM.2021. 3087038

    2022 doi

  50. [58]

    Yisen Wang, Xingjun Ma, James Bailey, Jinfeng Yi, Bowen Zhou, and Quanquan Gu. 2021. On the convergence and robustness of adversarial training. arXiv preprint arXiv:2112.08304 (2021)

    2021 arXiv

  51. [59]

    Eyal Winter. 2002. The shapley value. Handbook of game theory with economic applications 3 (2002), 2025–2054

    2002

  52. [60]

    Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, Rémi Louf, Morgan Funtowicz, et al

  53. [61]

    Chen Wu, Ruqing Zhang, Jiafeng Guo, Wei Chen, Yixing Fan, Maarten de Rijke, and Xueqi Cheng. 2022. Certified Robustness to Word Substitution Ranking Attack for Neural Ranking Models. In Proceedings of the 31st ACM International Conference on Information & Knowledge Management....

    2022

  54. [62]

    Chen Wu, Ruqing Zhang, Jiafeng Guo, Maarten de Rijke, Yixing Fan, and Xueqi Cheng. 2023. PRADA: Practical Black-box Adversarial Attacks against Neural Ranking Models. ACM Trans. Inf. Syst. 41, 4 (2023), 89:1–89:27. https://doi.org/ 10.1145/3576923

    2023 doi

  55. [63]

    arXiv preprint arXiv:1910.03771 (2019)

    Huggingface’s transformers: State-of-the-art natural language processing. arXiv preprint arXiv:1910.03771 (2019)

    2019 arXiv

  56. [64]

    Chenyan Xiong, Zhuyun Dai, Jamie Callan, Zhiyuan Liu, and Russell Power

  57. [65]

    Bennett, Junaid Ahmed, and Arnold Overwijk

    Lee Xiong, Chenyan Xiong, Ye Li, Kwok-Fung Tang, Jialin Liu, Paul N. Bennett, Junaid Ahmed, and Arnold Overwijk. 2021. Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval. In 9th International Conference on Learning Representations, ICLR 2021, V...

    2021

  58. [66]

    Chen Wu, Ruqing Zhang, Jiafeng Guo, Maarten De Rijke, Yixing Fan, and Xueqi Cheng. 2023. Prada: practical black-box adversarial attacks against neural ranking models. ACM Transactions on Information Systems 41, 4 (2023), 1–27

    2023

  59. [67]

    Wenhan Xiong, Xiang Lorraine Li, Srini Iyer, Jingfei Du, Patrick S. H. Lewis, William Yang Wang, Yashar Mehdad, Scott Yih, Sebastian Riedel, Douwe Kiela, and Barlas Oguz. 2021. Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval. In 9th International Confere...

    2021

  60. [68]

    Mao Ye, Chengyue Gong, and Qiang Liu. 2020. SAFER: A structure-free ap- proach for certified robustness to adversarial word substitutions. arXiv preprint arXiv:2005.14424 (2020)

    2020 arXiv

  61. [70]

    Lee Xiong, Chenyan Xiong, Ye Li, Kwok-Fung Tang, Jialin Liu, Paul Bennett, Junaid Ahmed, and Arnold Overwijk. 2020. Approximate nearest neighbor nega- tive contrastive learning for dense text retrieval. arXiv preprint arXiv:2007.00808 (2020). Unsupervised dense retrieval with ...

    2020 arXiv

  62. [2017]

    End-to-End Neural Ad-hoc Ranking with Kernel Pooling. In Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval, Shinjuku, Tokyo, Japan, August 7-11, 2017, Noriko Kando, Tetsuya Sakai, Hideo Joho, Hang Li, Arjen P. de Vr...

    2017

  63. [2019]

    In Proceedings of the 36th Interna- tional Conference on Machine Learning, ICML 2019, 9-15 June 2019, Long Beach, California, USA (Proceedings of Machine Learning Research, Vol

    Counterfactual Visual Explanations. In Proceedings of the 36th Interna- tional Conference on Machine Learning, ICML 2019, 9-15 June 2019, Long Beach, California, USA (Proceedings of Machine Learning Research, Vol. 97), Kamalika Chaudhuri and Ruslan Salakhutdinov (Eds.). PMLR, ...

    2019

  64. [2020]

    In Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval, SIGIR 2020, Virtual Event, China, July 25-30, 2020 , Jimmy X

    Open-Retrieval Conversational Question Answering. In Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval, SIGIR 2020, Virtual Event, China, July 25-30, 2020 , Jimmy X. Huang, Yi Chang, Xueqi Cheng, Jaap Kamps, Vanessa...

    2020

Pith tools

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