# Informativeness and Invariance: Two Perspectives on Spurious Correlations in Natural Language

arXiv: https://arxiv.org/abs/2204.04487 · HTML: https://mlsanity.com/paper/2204.04487/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 22 references that are in the map.
- First posted on arXiv 9 Apr 2022.

## Builds on (most cited first, 12 of 12)

- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [Hypothesis Only Baselines in Natural Language Inference](https://mlsanity.com/paper/1805.01042/) (2018)
- [Gender Bias in Neural Natural Language Processing](https://mlsanity.com/paper/1807.11714/) (2018)
- [Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyond](https://mlsanity.com/paper/2109.00725/) (2021)
- [Invariance, Causality and Robustness](https://mlsanity.com/paper/1812.08233/) (2018)
- [Competency Problems: On Finding and Removing Artifacts in Language Data](https://mlsanity.com/paper/2104.08646/) (2021)
- [Explaining The Efficacy of Counterfactually Augmented Data](https://mlsanity.com/paper/2010.02114/) (2020)
- [Causal Direction of Data Collection Matters: Implications of Causal and Anticausal Learning for NLP](https://mlsanity.com/paper/2110.03618/) (2021)

## Built on by (most cited first, 21 of 21)

- [Reasoning or Reciting? Exploring the Capabilities and Limitations of Language Models Through Counterfactual Tasks](https://mlsanity.com/paper/2307.02477/) (2023)
- [On Feature Learning in the Presence of Spurious Correlations](https://mlsanity.com/paper/2210.11369/) (2022)
- [Measuring Causal Effects of Data Statistics on Language Model's `Factual' Predictions](https://mlsanity.com/paper/2207.14251/) (2022)
- [NeuroCounterfactuals: Beyond Minimal-Edit Counterfactuals for Richer Data Augmentation](https://mlsanity.com/paper/2210.12365/) (2022)
- [Nuisances via Negativa: Adjusting for Spurious Correlations via Data Augmentation](https://mlsanity.com/paper/2210.01302/) (2022)
- [Finding Dataset Shortcuts with Grammar Induction](https://mlsanity.com/paper/2210.11560/) (2022)
- [Shift-Robust Molecular Relational Learning with Causal Substructure](https://mlsanity.com/paper/2305.18451/) (2023)
- [AGRO: Adversarial Discovery of Error-prone groups for Robust Optimization](https://mlsanity.com/paper/2212.00921/) (2022)
- [Robust Natural Language Understanding with Residual Attention Debiasing](https://mlsanity.com/paper/2305.17627/) (2023)
- [CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation](https://mlsanity.com/paper/2306.00374/) (2023)
- [RNR: Teaching Large Language Models to Follow Roles and Rules](https://mlsanity.com/paper/2409.13733/) (2024)
- [Establishing Trustworthiness: Rethinking Tasks and Model Evaluation](https://mlsanity.com/paper/2310.05442/) (2023)
- [Stubborn Lexical Bias in Data and Models](https://mlsanity.com/paper/2306.02190/) (2023)
- [Understanding Text Classification Data and Models Using Aggregated Input Salience](https://mlsanity.com/paper/2211.05485/) (2022)
- [CASCADE Your Datasets for Cross-Mode Knowledge Retrieval of Language Models](https://mlsanity.com/paper/2504.01450/) (2025)
- [The Distributional Hypothesis Does Not Fully Explain the Benefits of Masked Language Model Pretraining](https://mlsanity.com/paper/2310.16261/) (2023)
- [Measuring and Improving Attentiveness to Partial Inputs with Counterfactuals](https://mlsanity.com/paper/2311.09605/) (2023)
- [Not Eliminate but Aggregate: Post-Hoc Control over Mixture-of-Experts to Address Shortcut Shifts in Natural Language Understanding](https://mlsanity.com/paper/2406.12060/) (2024)
- [The Pragmatic Frames of Spurious Correlations in Machine Learning: Interpreting How and Why They Matter](https://mlsanity.com/paper/2411.04696/) (2024)
- [Counterfactual Samples Constructing and Training for Commonsense Statements Estimation](https://mlsanity.com/paper/2412.20563/) (2024)
- [Exposing the Unsaid: Visualizing Hidden LLM Bias through Stochastic Path Aggregation](https://mlsanity.com/paper/2606.19344/) (2026)

## Abstract (arXiv)

Spurious correlations are a threat to the trustworthiness of natural language processing systems, motivating research into methods for identifying and eliminating them. However, addressing the problem of spurious correlations requires more clarity on what they are and how they arise in language data. Gardner et al (2021) argue that the compositional nature of language implies that all correlations between labels and individual "input features" are spurious. This paper analyzes this proposal in the context of a toy example, demonstrating three distinct conditions that can give rise to feature-label correlations in a simple PCFG. Linking the toy example to a structured causal model shows that (1) feature-label correlations can arise even when the label is invariant to interventions on the feature, and (2) feature-label correlations may be absent even when the label is sensitive to interventions on the feature. Because input features will be individually correlated with labels in all but very rare circumstances, domain knowledge must be applied to identify spurious correlations that pose genuine robustness threats.