Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
102 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

NatLogAttack: A Framework for Attacking Natural Language Inference Models with Natural Logic (2307.02849v2)

Published 6 Jul 2023 in cs.CL

Abstract: Reasoning has been a central topic in artificial intelligence from the beginning. The recent progress made on distributed representation and neural networks continues to improve the state-of-the-art performance of natural language inference. However, it remains an open question whether the models perform real reasoning to reach their conclusions or rely on spurious correlations. Adversarial attacks have proven to be an important tool to help evaluate the Achilles' heel of the victim models. In this study, we explore the fundamental problem of developing attack models based on logic formalism. We propose NatLogAttack to perform systematic attacks centring around natural logic, a classical logic formalism that is traceable back to Aristotle's syllogism and has been closely developed for natural language inference. The proposed framework renders both label-preserving and label-flipping attacks. We show that compared to the existing attack models, NatLogAttack generates better adversarial examples with fewer visits to the victim models. The victim models are found to be more vulnerable under the label-flipping setting. NatLogAttack provides a tool to probe the existing and future NLI models' capacity from a key viewpoint and we hope more logic-based attacks will be further explored for understanding the desired property of reasoning.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (72)
  1. Generating natural language adversarial examples. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 2890–2896, Brussels, Belgium. Association for Computational Linguistics.
  2. Gabor Angeli and Christopher D Manning. 2014. Naturalli: Natural logic inference for common sense reasoning. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), Doha, Qatar.
  3. Combining natural logic and shallow reasoning for question answering. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (ACL), Berlin, Germany.
  4. Don’t take the premise for granted: Mitigating artifacts in natural language inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL), pages 877–891.
  5. A large annotated corpus for learning natural language inference. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 632–642, Lisbon, Portugal. Association for Computational Linguistics.
  6. Neural natural language inference models enhanced with external knowledge. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (ACL), Melbourne, Australia.
  7. Enhanced lstm for natural language inference. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (ACL), Vancouver. ACL.
  8. Zeming Chen and Qiyue Gao. 2021. Monotonicity marking from universal dependency trees. In Proceedings of the 14th International Conference on Computational Semantics (IWCS), pages 121–131.
  9. Neurallog: Natural language inference with joint neural and logical reasoning. In Proceedings of* SEM 2021: The Tenth Joint Conference on Lexical and Computational Semantics, pages 78–88.
  10. The PASCAL recognising textual entailment challenge. In Proceedings of the First international conference on Machine Learning Challenges: evaluating Predictive Uncertainty Visual Object Classification, and Recognizing Textual Entailment.
  11. Neuro-symbolic= neural+ logical+ probabilistic. In NeSy’19@ IJCAI, the 14th International Workshop on Neural-Symbolic Learning and Reasoning, Macao, China.
  12. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 4171–4186, Minneapolis, Minnesota. Association for Computational Linguistics.
  13. HowNet and its computation of meaning. In Coling 2010: Demonstrations, pages 53–56, Beijing, China. Coling 2010 Organizing Committee.
  14. HotFlip: White-box adversarial examples for text classification. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 31–36, Melbourne, Australia. Association for Computational Linguistics.
  15. Richard Evans and Edward Grefenstette. 2018. Learning explanatory rules from noisy data. In Journal of Artificial Intelligence Research (JAIR), volume 61, pages 1–64.
  16. Neuro-symbolic natural logic with introspective revision for natural language inference. Transactions of the Association for Computational Linguistics (TACL), 10:240–256.
  17. Exploring end-to-end differentiable natural logic modeling. In Proceedings of the 28th International Conference on Computational Linguistics (COLING), pages 1172–1185.
  18. Neural-symbolic learning and reasoning: contributions and challenges. In 2015 AAAI Spring Symposium Series.
  19. Neural natural language inference models partially embed theories of lexical entailment and negation. In Proceedings of the Third BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, pages 163–173.
  20. Breaking NLI systems with sentences that require simple lexical inferences. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 650–655, Melbourne, Australia. Association for Computational Linguistics.
  21. Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572.
  22. Monalog: a lightweight system for natural language inference based on monotonicity.
  23. Hai Hu and Larry Moss. 2018. Polarity computations in flexible categorial grammar. In Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics, pages 124–129.
  24. Thomas F Icard. 2012. Inclusion and exclusion in natural language. Studia Logica.
  25. Thomas F Icard and Lawrence S Moss. 2014. Recent progress on monotonicity. In Linguistic Issues in Language Technology. Citeseer.
  26. Adrian Iftene and Alexandra Balahur-Dobrescu. 2007. Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing. Prague, Czech.
  27. Adversarial example generation with syntactically controlled paraphrase networks. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), pages 1875–1885.
  28. Robin Jia and Percy Liang. 2017. Adversarial examples for evaluating reading comprehension systems. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 2021–2031, Copenhagen, Denmark. Association for Computational Linguistics.
  29. Is bert really robust? a strong baseline for natural language attack on text classification and entailment. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 8018–8025.
  30. Adventure: Adversarial training for textual entailment with knowledge-guided examples. In The 56th Annual Meeting of the Association for Computational Linguistics (ACL), Melbourne, Australia.
  31. Adversarial examples in the physical world. In Artificial intelligence safety and security, pages 99–112. Chapman and Hall/CRC.
  32. George Lakoff. 1970. Linguistics and natural logic. Synthese, 22(1-2):151–271.
  33. Contextualized perturbation for textual adversarial attack. In Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics (NAACL).
  34. BERT-ATTACK: Adversarial attack against BERT using BERT. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 6193–6202, Online. Association for Computational Linguistics.
  35. Deep text classification can be fooled. In Proceedings of the 27th International Joint Conference on Artificial Intelligence (IJCAI), pages 4208–4215.
  36. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692.
  37. Bill MacCartney. 2009. Natural Language Inference. Ph.D. thesis, Stanford University.
  38. Bill MacCartney and Christopher D Manning. 2009. An extended model of natural logic. In Proceedings of the 8th international conference on computational semantics (IWCS), Stroudsburg, United States.
  39. The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervision. In Proceedings of the 7th International Conference on Learning Representations (ICLR), New Orleans, USA.
  40. A sick cure for the evaluation of compositional distributional semantic models. In Proceedings of the 9th International Conference on Language Resources and Evaluation (LREC), Reykjavik, Iceland.
  41. Right for the wrong reasons: Diagnosing syntactic heuristics in natural language inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL), pages 3428–3448.
  42. George A Miller. 1995. Wordnet: a lexical database for english. Communications of the ACM, 38(11):39–41.
  43. Pasquale Minervini and Sebastian Riedel. 2018. Adversarially regularising neural nli models to integrate logical background knowledge. In Proceedings of the 22nd Conference on Computational Natural Language Learning (CoNLL), pages 65–74.
  44. Textattack: A framework for adversarial attacks, data augmentation, and adversarial training in nlp. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 119–126.
  45. Computing relative polarity for textual inference. In Proceedings of the 5th international workshop on inference in computational semantics, Buxton, England.
  46. Barbara Partee. 1995. Lexical semantics and compositionality. Invitation to Cognitive Science.
  47. Conditionally adaptive multi-task learning: Improving transfer learning in nlp using fewer parameters & less data. In International Conference on Learning Representations (ICLR).
  48. Collecting diverse natural language inference problems for sentence representation evaluation. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 67–81, Brussels, Belgium. Association for Computational Linguistics.
  49. Language models are unsupervised multitask learners. OpenAI Blog, 1(8):9.
  50. Generating natural language adversarial examples through probability weighted word saliency. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL), pages 1085–1097.
  51. Probing natural language inference models through semantic fragments. In Proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI), New York, USA.
  52. Tim Rocktäschel and Sebastian Riedel. 2017. End-to-end differentiable proving. In Proceedings of the 31st International Conference on Neural Information Processing Systems (NeurIPS), Long Beach, USA.
  53. Masked language model scoring. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 2699–2712, Online. Association for Computational Linguistics.
  54. Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter. the 5th Workshop on Energy Efficient Machine Learning and Cognitive Computing @NeurIPS.
  55. Interpretable adversarial perturbation in input embedding space for text. In Proceedings of the 27th International Joint Conference on Artificial Intelligence (IJCAI), pages 4323–4330.
  56. Fundamental tradeoffs between invariance and sensitivity to adversarial perturbations. In International Conference on Machine Learning, pages 9561–9571. PMLR.
  57. Víctor Manuel Sánchez Valencia. 1991. Studies on natural logic and categorial grammar. Universiteit van Amsterdam.
  58. Johan van Benthem. 1988. The semantics of variety in categorial grammar. Categorial grammar.
  59. Johan Van Benthem. 1995. Language in Action: categories, lambdas and dynamic logic. MIT Press.
  60. Johan Van Benthem et al. 1986. Essays in logical semantics. Springer.
  61. Universal adversarial triggers for attacking and analyzing NLP. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 2153–2162, Hong Kong, China. Association for Computational Linguistics.
  62. Nlprolog: Reasoning with weak unification for question answering in natural language. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL), Austin, Texas, United States.
  63. A broad-coverage challenge corpus for sentence understanding through inference. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), pages 1112–1122, New Orleans, Louisiana. Association for Computational Linguistics.
  64. Do neural models learn systematicity of monotonicity inference in natural language? In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL), pages 6105–6117.
  65. Can neural networks understand monotonicity reasoning? In Proceedings of the 2019 ACL Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, Austin, Texas, United States.
  66. Help: A dataset for identifying shortcomings of neural models in monotonicity reasoning. In Proceedings of the Eighth Joint Conference on Lexical and Computational Semantics (*SEM), Minneapolis, Minnesota, USA.
  67. Differentiable learning of logical rules for knowledge base reasoning. In Advances in Neural Information Processing Systems, pages 2319–2328.
  68. Word-level textual adversarial attacking as combinatorial optimization. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL), pages 6066–6080.
  69. Openattack: An open-source textual adversarial attack toolkit. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing: System Demonstrations, pages 363–371.
  70. Generating fluent adversarial examples for natural languages. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL), pages 5564–5569, Florence, Italy. Association for Computational Linguistics.
  71. Semantics-aware bert for language understanding. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 9628–9635.
  72. Generating natural adversarial examples. In Proceedings of the 6th International Conference on Learning Representations (ICLR).
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (2)
  1. Zi'ou Zheng (4 papers)
  2. Xiaodan Zhu (94 papers)
Citations (3)
View on Semantic Scholar