Papers
Topics
Authors
Recent
Search
2000 character limit reached

Explore Spurious Correlations at the Concept Level in Language Models for Text Classification

Published 15 Nov 2023 in cs.CL and cs.AI | (2311.08648v4)

Abstract: LLMs (LMs) have achieved notable success in numerous NLP tasks, employing both fine-tuning and in-context learning (ICL) methods. While LLMs demonstrate exceptional performance, they face robustness challenges due to spurious correlations arising from imbalanced label distributions in training data or ICL exemplars. Previous research has primarily concentrated on word, phrase, and syntax features, neglecting the concept level, often due to the absence of concept labels and difficulty in identifying conceptual content in input texts. This paper introduces two main contributions. First, we employ ChatGPT to assign concept labels to texts, assessing concept bias in models during fine-tuning or ICL on test data. We find that LMs, when encountering spurious correlations between a concept and a label in training or prompts, resort to shortcuts for predictions. Second, we introduce a data rebalancing technique that incorporates ChatGPT-generated counterfactual data, thereby balancing label distribution and mitigating spurious correlations. Our method's efficacy, surpassing traditional token removal approaches, is validated through extensive testing.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (55)
  1. Cebab: Estimating the causal effects of real-world concepts on nlp model behavior. Advances in Neural Information Processing Systems, 35:17582–17596.
  2. Generating natural language adversarial examples. arXiv preprint arXiv:1804.07998.
  3. Fast optimal leaf ordering for hierarchical clustering. Bioinformatics, 17(suppl_1):S22–S29.
  4. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901.
  5. Understanding and mitigating spurious correlations in text classification. arXiv preprint arXiv:2305.13654.
  6. Learning to model and ignore dataset bias with mixed capacity ensembles. arXiv preprint arXiv:2011.03856.
  7. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
  8. Yanbo Fang and Yongfeng Zhang. 2022. Data-efficient concept extraction from pre-trained language models for commonsense explanation generation. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 5883–5893.
  9. Making pre-trained language models better few-shot learners. arXiv preprint arXiv:2012.15723.
  10. Shortcut learning in deep neural networks. Nature Machine Intelligence, 2(11):665–673.
  11. Chatgpt outperforms crowd-workers for text-annotation tasks. arXiv preprint arXiv:2303.15056.
  12. Unlearn dataset bias in natural language inference by fitting the residual. arXiv preprint arXiv:1908.10763.
  13. Ruining He and Julian McAuley. 2016. Ups and downs: Modeling the visual evolution of fashion trends with one-class collaborative filtering. In proceedings of the 25th international conference on world wide web, pages 507–517.
  14. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685.
  15. Robin Jia and Percy Liang. 2017. Adversarial examples for evaluating reading comprehension systems. arXiv preprint arXiv:1707.07328.
  16. 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.
  17. Learning the difference that makes a difference with counterfactually-augmented data. arXiv preprint arXiv:1909.12434.
  18. Algorithmic fairness. In Aea papers and proceedings, volume 108, pages 22–27. American Economic Association 2014 Broadway, Suite 305, Nashville, TN 37203.
  19. Why machine reading comprehension models learn shortcuts? In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 989–1002, Online. Association for Computational Linguistics.
  20. Adversarial filters of dataset biases. In International conference on machine learning, pages 1078–1088. PMLR.
  21. Just train twice: Improving group robustness without training group information. In International Conference on Machine Learning, pages 6781–6792. PMLR.
  22. Pre-train, prompt, and predict: A systematic survey of prompting methods in natural language processing. ACM Computing Surveys, 55(9):1–35.
  23. Cat: Causal audio transformer for audio classification. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE.
  24. C-disentanglement: Discovering causally-independent generative factors under an inductive bias of confounder. arXiv preprint arXiv:2310.17325.
  25. Fantastically ordered prompts and where to find them: Overcoming few-shot prompt order sensitivity. arXiv preprint arXiv:2104.08786.
  26. Learning word vectors for sentiment analysis. In Proceedings of the 49th annual meeting of the association for computational linguistics: Human language technologies, pages 142–150.
  27. Right for the wrong reasons: Diagnosing syntactic heuristics in natural language inference. arXiv preprint arXiv:1902.01007.
  28. Automatic shortcut removal for self-supervised representation learning. In International Conference on Machine Learning, pages 6927–6937. PMLR.
  29. Evaluating robustness to input perturbations for neural machine translation. arXiv preprint arXiv:2005.00580.
  30. OpenAI. 2023. Gpt-4 technical report.
  31. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35:27730–27744.
  32. Hypothesis only baselines in natural language inference. In Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics, pages 180–191.
  33. Beyond accuracy: Behavioral testing of nlp models with checklist. arXiv preprint arXiv:2005.04118.
  34. An investigation of why overparameterization exacerbates spurious correlations. In International Conference on Machine Learning, pages 8346–8356. PMLR.
  35. Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108.
  36. Timo Schick and Hinrich Schütze. 2020. Exploiting cloze questions for few shot text classification and natural language inference. arXiv preprint arXiv:2001.07676.
  37. Tackling the story ending biases in the story cloze test. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 752–757.
  38. Measuring inductive biases of in-context learning with underspecified demonstrations. arXiv preprint arXiv:2305.13299.
  39. Large language models can be lazy learners: Analyze shortcuts in in-context learning. arXiv preprint arXiv:2305.17256.
  40. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288.
  41. An empirical study on robustness to spurious correlations using pre-trained language models. Transactions of the Association for Computational Linguistics, 8:621–633.
  42. Mind the trade-off: Debiasing nlu models without degrading the in-distribution performance. arXiv preprint arXiv:2005.00315.
  43. Identifying and mitigating spurious correlations for improving robustness in NLP models. In Findings of the Association for Computational Linguistics: NAACL 2022, pages 1719–1729, Seattle, United States. Association for Computational Linguistics.
  44. Chain-of-thought prompting elicits reasoning in large language models. Advances in Neural Information Processing Systems, 35:24824–24837.
  45. Robert Wolfe and Aylin Caliskan. 2021. Low frequency names exhibit bias and overfitting in contextualizing language models. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 518–532, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
  46. Hellaswag: Can a machine really finish your sentence? arXiv preprint arXiv:1905.07830.
  47. Correct-n-contrast: A contrastive approach for improving robustness to spurious correlations. arXiv preprint arXiv:2203.01517.
  48. Character-level convolutional networks for text classification. Advances in neural information processing systems, 28.
  49. Gender bias in coreference resolution: Evaluation and debiasing methods. arXiv preprint arXiv:1804.06876.
  50. Investigating ensemble methods for model robustness improvement of text classifiers. arXiv preprint arXiv:2210.16298.
  51. Calibrate before use: Improving few-shot performance of language models. In International Conference on Machine Learning, pages 12697–12706. PMLR.
  52. On large language models’ selection bias in multi-choice questions. arXiv preprint arXiv:2309.03882.
  53. Xiang Zhou and Mohit Bansal. 2020. Towards robustifying NLI models against lexical dataset biases. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 8759–8771, Online. Association for Computational Linguistics.
  54. Scalable prompt generation for semi-supervised learning with language models. In Findings of the Association for Computational Linguistics: EACL 2023, pages 758–769.
  55. Spottarget: Rethinking the effect of target edges for link prediction in graph neural networks. arXiv preprint arXiv:2306.00899.
Citations (17)
View on Semantic Scholar

Summary

No one has generated a summary of this paper yet.

Sign Up to Summarize

Paper to Video (Beta)

No one has generated a video about this paper yet.

Sign Up to Generate All Videos Subscribe on YouTube

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Generate Now

Tweets

Sign up for free to view the 2 tweets with 55 likes about this paper.

Sign Up for Free