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

arXiv: https://arxiv.org/abs/2311.08648 · HTML: https://mlsanity.com/paper/2311.08648/

## Key facts

- Cited by 48 papers in the map (data through 14 Oct 2026).
- Builds on 48 of its 73 references that are in the map.
- First posted on arXiv 15 Nov 2023.

## Builds on (most cited first, 30 of 48)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Calibrate Before Use: Improving Few-Shot Performance of Language Models](https://mlsanity.com/paper/2102.09690/) (2021)
- [Making Pre-trained Language Models Better Few-shot Learners](https://mlsanity.com/paper/2012.15723/) (2020)
- [Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity](https://mlsanity.com/paper/2104.08786/) (2021)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference](https://mlsanity.com/paper/2001.07676/) (2020)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks](https://mlsanity.com/paper/2303.15056/) (2023)
- [Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment](https://mlsanity.com/paper/1907.11932/) (2019)
- [Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods](https://mlsanity.com/paper/1804.06876/) (2018)
- [Adversarial Examples for Evaluating Reading Comprehension Systems](https://mlsanity.com/paper/1707.07328/) (2017)
- [Just Train Twice: Improving Group Robustness without Training Group Information](https://mlsanity.com/paper/2107.09044/) (2021)
- [Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering](https://mlsanity.com/paper/1602.01585/) (2016)
- [Learning the Difference that Makes a Difference with Counterfactually-Augmented Data](https://mlsanity.com/paper/1909.12434/) (2019)
- [Generating Natural Language Adversarial Examples](https://mlsanity.com/paper/1804.07998/) (2018)
- [An Investigation of Why Overparameterization Exacerbates Spurious Correlations](https://mlsanity.com/paper/2005.04345/) (2020)
- [Hypothesis Only Baselines in Natural Language Inference](https://mlsanity.com/paper/1805.01042/) (2018)
- [Correct-N-Contrast: A Contrastive Approach for Improving Robustness to Spurious Correlations](https://mlsanity.com/paper/2203.01517/) (2022)
- [Adversarial Filters of Dataset Biases](https://mlsanity.com/paper/2002.04108/) (2020)
- [Algorithmic Fairness](https://mlsanity.com/paper/2001.09784/) (2020)
- [An Empirical Study on Robustness to Spurious Correlations using Pre-trained Language Models](https://mlsanity.com/paper/2007.06778/) (2020)

## Built on by (most cited first, 30 of 48)

- [Mementos: A Comprehensive Benchmark for Multimodal Large Language Model Reasoning over Image Sequences](https://mlsanity.com/paper/2401.10529/) (2024)
- [Enhancing Visual-Language Modality Alignment in Large Vision Language Models via Self-Improvement](https://mlsanity.com/paper/2405.15973/) (2024)
- [The Best Instruction-Tuning Data are Those That Fit](https://mlsanity.com/paper/2502.04194/) (2025)
- [Which of These Best Describes Multiple Choice Evaluation with LLMs? A) Forced B) Flawed C) Fixable D) All of the Above](https://mlsanity.com/paper/2502.14127/) (2025)
- [Inoculation Prompting: Instructing LLMs to misbehave at train-time improves test-time alignment](https://mlsanity.com/paper/2510.05024/) (2025)
- [Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment](https://mlsanity.com/paper/2501.09620/) (2025)
- [The Clever Hans Mirage: A Comprehensive Survey on Spurious Correlations in Machine Learning](https://mlsanity.com/paper/2402.12715/) (2024)
- [Large Language Models and Causal Inference in Collaboration: A Survey](https://mlsanity.com/paper/2403.09606/) (2024)
- [A Survey on Natural Language Counterfactual Generation](https://mlsanity.com/paper/2407.03993/) (2024)
- [CausalChaos! Dataset for Comprehensive Causal Action Question Answering Over Longer Causal Chains Grounded in Dynamic Visual Scenes](https://mlsanity.com/paper/2404.01299/) (2024)
- [MergeME: Model Merging Techniques for Homogeneous and Heterogeneous MoEs](https://mlsanity.com/paper/2502.00997/) (2025)
- [Shortcut Learning in In-Context Learning: A Survey](https://mlsanity.com/paper/2411.02018/) (2024)
- [Emojis Decoded: Leveraging ChatGPT for Enhanced Understanding in Social Media Communications](https://mlsanity.com/paper/2402.01681/) (2024)
- [DISCO Balances the Scales: Adaptive Domain- and Difficulty-Aware Reinforcement Learning on Imbalanced Data](https://mlsanity.com/paper/2505.15074/) (2025)
- [BOAD: Discovering Hierarchical Software Engineering Agents via Bandit Optimization](https://mlsanity.com/paper/2512.23631/) (2025)
- [Evaluate Bias without Manual Test Sets: A Concept Representation Perspective for LLMs](https://mlsanity.com/paper/2505.15524/) (2025)
- [Stochastic Chameleons: Irrelevant Context Hallucinations Reveal Class-Based (Mis)Generalization in LLMs](https://mlsanity.com/paper/2505.22630/) (2025)
- [Rectifying Shortcut Behaviors in Preference-based Reward Learning](https://mlsanity.com/paper/2510.19050/) (2025)
- [Debiasing Reward Models by Representation Learning with Guarantees](https://mlsanity.com/paper/2510.23751/) (2025)
- [Mitigating Biases of Large Language Models in Stance Detection with Counterfactual Augmented Calibration](https://mlsanity.com/paper/2402.14296/) (2024)
- [Multi-Stage Balanced Distillation: Addressing Long-Tail Challenges in Sequence-Level Knowledge Distillation](https://mlsanity.com/paper/2406.13114/) (2024)
- [Focus On This, Not That! Steering LLMs with Adaptive Feature Specification](https://mlsanity.com/paper/2410.22944/) (2024)
- [CSRec: Rethinking Sequential Recommendation from A Causal Perspective](https://mlsanity.com/paper/2409.05872/) (2024)
- [Retrieving Counterfactuals Improves Visual In-Context Learning](https://mlsanity.com/paper/2603.16737/) (2026)
- [Skill Discovery for Software Scripting Automation via Offline Simulations with LLMs](https://mlsanity.com/paper/2504.20406/) (2025)
- [When Bias Pretends to Be Truth: How Spurious Correlations Undermine Hallucination Detection in LLMs](https://mlsanity.com/paper/2511.07318/) (2025)
- [Exploring Explanations Improves the Robustness of In-Context Learning](https://mlsanity.com/paper/2506.02378/) (2025)
- [LoRA Users Beware: A Few Spurious Tokens Can Manipulate Your Finetuned Model](https://mlsanity.com/paper/2506.11402/) (2025)
- [Consistency Training Can Entrench Misalignment](https://mlsanity.com/paper/2606.03810/) (2026)
- [From Adoption to Adaption: Tracing the Diffusion of New Emojis on Twitter](https://mlsanity.com/paper/2402.14187/) (2024)

## Abstract (arXiv)

Language models (LMs) have achieved notable success in numerous NLP tasks, employing both fine-tuning and in-context learning (ICL) methods. While language models 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.