English

Mitigating Spurious Correlations in LLMs via Causality-Aware Post-Training

Machine Learning 2025-06-12 v1

Abstract

While large language models (LLMs) have demonstrated remarkable capabilities in language modeling, recent studies reveal that they often fail on out-of-distribution (OOD) samples due to spurious correlations acquired during pre-training. Here, we aim to mitigate such spurious correlations through causality-aware post-training (CAPT). By decomposing a biased prediction into two unbiased steps, known as \textit{event estimation} and \textit{event intervention}, we reduce LLMs' pre-training biases without incurring additional fine-tuning biases, thus enhancing the model's generalization ability. Experiments on the formal causal inference benchmark CLadder and the logical reasoning dataset PrOntoQA show that 3B-scale language models fine-tuned with CAPT can outperform both traditional SFT and larger LLMs on in-distribution (ID) and OOD tasks using only 100 ID fine-tuning samples, demonstrating the effectiveness and sample efficiency of CAPT.

Keywords

Cite

@article{arxiv.2506.09433,
  title  = {Mitigating Spurious Correlations in LLMs via Causality-Aware Post-Training},
  author = {Shurui Gui and Shuiwang Ji},
  journal= {arXiv preprint arXiv:2506.09433},
  year   = {2025}
}
R2 v1 2026-07-01T03:10:39.491Z