Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
97 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Competitive Retrieval: Going Beyond the Single Query (2404.09253v1)

Published 14 Apr 2024 in cs.IR and cs.GT

Abstract: Previous work on the competitive retrieval setting focused on a single-query setting: document authors manipulate their documents so as to improve their future ranking for a given query. We study a competitive setting where authors opt to improve their document's ranking for multiple queries. We use game theoretic analysis to prove that equilibrium does not necessarily exist. We then empirically show that it is more difficult for authors to improve their documents' rankings for multiple queries with a neural ranker than with a state-of-the-art feature-based ranker. We also present an effective approach for predicting the document most highly ranked in the next induced ranking.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (44)
  1. UQV100: A Test Collection with Query Variability. In Proceedings of SIGIR. 465–474.
  2. The Probability Ranking Principle is Not Optimal in Adversarial Retrieval Settings. In Proceedings of ICTIR. 51–60.
  3. A Game Theoretic Analysis of the Adversarial Retrieval Setting. J. Artif. Intell. Res. 60 (2017), 1127–1164.
  4. Quality-biased ranking of web documents. In Proceedings of WSDM. 95–104.
  5. Parameterized concept weighting in verbose queries. In Proceedings of SIGIR. 605–614.
  6. Carlos Castillo and Brian D. Davison. 2010. Adversarial Web Search. Foundations and Trends in Information Retrieval 4, 5 (2010), 377–486.
  7. Defense of Adversarial Ranking Attack in Text Retrieval: Benchmark and Baseline via Detection. CoRR abs/2307.16816 (2023).
  8. Towards Imperceptible Document Manipulations against Neural Ranking Models. In Findings of the Association for Computational Linguistics: ACL. 6648–6664.
  9. Reciprocal rank fusion outperforms condorcet and individual rank learning methods. In Proceedings of SIGIR. 758–759.
  10. Focused Retrieval of Content and Structure, 10th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2011, Saarbrücken, Germany, December 12-14, 2011, Revised Selected Papers.
  11. Ranking Robustness Under Adversarial Document Manipulations. In Proceedings of SIGIR. 395–404.
  12. Ranking-Incentivized Quality Preserving Content Modification. In Proceedings of SIGIR. 259–268.
  13. Driving the Herd: Search Engines as Content Influencers. In Proceedings of SIGIR. 586–595.
  14. Zoltán Gyöngyi and Hector Garcia-Molina. 2005. Web Spam Taxonomy. In Proceedings of AIRWeb 2005, First International Workshop on Adversarial Information Retrieval on the Web. 39–47.
  15. Introducing Connect by CloudResearch: Advancing Online Participant Recruitment in the Digital Age.
  16. Adversarial Attacks on Neural Network Policies. In Proceedings of ICLR.
  17. Robin Jia and Percy Liang. 2017. Adversarial Examples for Evaluating Reading Comprehension Systems. In Proc. of EMNLP. 2021–2031.
  18. Certified Robustness to Adversarial Word Substitutions. In Proceedings of EMNLP-IJCNLP. 4127–4140.
  19. Accurately interpreting clickthrough data as implicit feedback. In Proceedings of SIGIR. 154–161.
  20. Fooling End-To-End Speaker Verification With Adversarial Examples. In Proceedings of ICASSP. 1962–1966.
  21. Oren Kurland and Moshe Tennenholtz. 2022. Competitive Search. In Proceedings of SIGIR. 2838–2849.
  22. Pretrained Transformers for Text Ranking: BERT and Beyond. CoRR abs/2010.06467 (2020).
  23. Order-Disorder: Imitation Adversarial Attacks for Black-box Neural Ranking Models. In Proceedings of SIGSAC. 2025–2039.
  24. Tie-Yan Liu. 2011. Learning to Rank for Information Retrieval. Springer. I–XVII, 1–285 pages.
  25. Topic-oriented Adversarial Attacks against Black-box Neural Ranking Models. In Proceedings of SIGIR. 1700–1709.
  26. John F. Nash. 1950. Equilibrium Points in N-Person Games. Proc. of the National Academy of Sciences of the United States of America 36.1 (1950), 48–49.
  27. Algorithmic Game Theory. Cambridge University Press. Section 19.3.
  28. Rodrigo Nogueira and Kyunghyun Cho. 2019. Passage Re-ranking with BERT. arXiv preprint arXiv:1901.04085 (2019).
  29. OpenAI. 2023. GPT-4 Technical Report. CoRR abs/2303.08774 (2023).
  30. Practical Black-Box Attacks against Machine Learning. In Proceedings of AsiaCCS. 506–519.
  31. Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research 12 (2011), 2825–2830.
  32. Information Retrieval Meets Game Theory: The Ranking Competition Between Documents’ Authors. In Proceedings of SIGIR. 465–474.
  33. Stephen E. Robertson. 1977. The Probability Ranking Principle in IR. Journal of Documentation (1977), 294–304. Reprinted in K. Sparck Jones and P. Willett (eds), Readings in Information Retrieval, pp. 281–286, 1997.
  34. TRAttack: Text Rewriting Attack Against Text Retrieval. In Proceedings of RepL4NLP@ACL. 191–203.
  35. Intriguing properties of neural networks. In Proceedings of ICLR.
  36. Content-Based Relevance Estimation in Retrieval Settings with Ranking-Incentivized Document Manipulations. In Proceedings of SIGIR. 205–2014.
  37. BERT Rankers are Brittle: A Study using Adversarial Document Perturbations. In Proceedings of ICTIR. 115–120.
  38. A similarity measure for indefinite rankings. ACM Trans. Inf. Syst. 28, 4 (2010), 20:1–20:38.
  39. PRADA: Practical Black-Box Adversarial Attacks against Neural Ranking Models. arXiv:2204.01321
  40. Are Neural Ranking Models Robust? ACM Trans. Inf. Syst. 41, 2 (2023), 29:1–29:36.
  41. Adapting boosting for information retrieval measures. Information Retrieval 13, 3 (2010), 254–270.
  42. Adversarial Examples for Semantic Segmentation and Object Detection. In Proceedings of ICCV 2017. 1378–1387.
  43. ChengXiang Zhai and John D. Lafferty. 2006. A risk minimization framework for information retrieval. Inf. Process. Manag. 42, 1 (2006), 31–55.
  44. Theoretically Principled Trade-off between Robustness and Accuracy. In Proceedings of ICML. 7472–7482.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (4)
  1. Haya Nachimovsky (5 papers)
  2. Moshe Tennenholtz (97 papers)
  3. Fiana Raiber (4 papers)
  4. Oren Kurland (17 papers)

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets