# SAGE: Spuriousness-Aware Guided Prompt Exploration for Mitigating Multimodal Bias

arXiv: https://arxiv.org/abs/2511.13005 · HTML: https://mlsanity.com/paper/2511.13005/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Prompt learning and adaptation of CLIP, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 26 of its 41 references that are in the map.
- First posted on arXiv 17 Nov 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Prompt learning and adaptation of CLIP](https://mlsanity.com/mllm/prompt-learning-and-adaptation-of-clip/)

## Builds on (most cited first, 26 of 26)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Align before Fuse: Vision and Language Representation Learning with Momentum Distillation](https://mlsanity.com/paper/2107.07651/) (2021)
- [Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs](https://mlsanity.com/paper/2401.06209/) (2024)
- [Aligning Large Multimodal Models with Factually Augmented RLHF](https://mlsanity.com/paper/2309.14525/) (2023)
- [Recognition in Terra Incognita](https://mlsanity.com/paper/1807.04975/) (2018)
- [Just Train Twice: Improving Group Robustness without Training Group Information](https://mlsanity.com/paper/2107.09044/) (2021)
- [Last Layer Re-Training is Sufficient for Robustness to Spurious Correlations](https://mlsanity.com/paper/2204.02937/) (2022)
- [Debiasing Vision-Language Models via Biased Prompts](https://mlsanity.com/paper/2302.00070/) (2023)
- [Spread Spurious Attribute: Improving Worst-group Accuracy with Spurious Attribute Estimation](https://mlsanity.com/paper/2204.02070/) (2022)
- [AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities](https://mlsanity.com/paper/2211.06679/) (2022)
- [Simple and Fast Group Robustness by Automatic Feature Reweighting](https://mlsanity.com/paper/2306.11074/) (2023)
- [Contrastive Adapters for Foundation Model Group Robustness](https://mlsanity.com/paper/2207.07180/) (2022)
- [Towards Last-layer Retraining for Group Robustness with Fewer Annotations](https://mlsanity.com/paper/2309.08534/) (2023)
- [Mitigating Spurious Correlations in Multi-modal Models during Fine-tuning](https://mlsanity.com/paper/2304.03916/) (2023)
- [Improving Zero-shot Generalization and Robustness of Multi-modal Models](https://mlsanity.com/paper/2212.01758/) (2022)
- [Robust Learning with Progressive Data Expansion Against Spurious Correlation](https://mlsanity.com/paper/2306.04949/) (2023)
- [MM-SpuBench: Towards Better Understanding of Spurious Biases in Multimodal LLMs](https://mlsanity.com/paper/2406.17126/) (2024)
- [Calibrating Multi-modal Representations: A Pursuit of Group Robustness without Annotations](https://mlsanity.com/paper/2403.07241/) (2024)
- [FairerCLIP: Debiasing CLIP's Zero-Shot Predictions using Functions in RKHSs](https://mlsanity.com/paper/2403.15593/) (2024)
- [Spuriousness-Aware Meta-Learning for Learning Robust Classifiers](https://mlsanity.com/paper/2406.10742/) (2024)
- [ShortcutProbe: Probing Prediction Shortcuts for Learning Robust Models](https://mlsanity.com/paper/2505.13910/) (2025)
- [Improving Group Robustness on Spurious Correlation via Evidential Alignment](https://mlsanity.com/paper/2506.11347/) (2025)
- [NeuronTune: Towards Self-Guided Spurious Bias Mitigation](https://mlsanity.com/paper/2505.24048/) (2025)

## Built on by (most cited first, 1 of 1)

- [When Masking Helps or Hurts Robustness in Compressed CLIP: A Pre-Deployment Diagnostic](https://mlsanity.com/paper/2609.39704/) (2026)

## Abstract (arXiv)

Large vision-language models, such as CLIP, have shown strong zero-shot classification performance by aligning images and text in a shared embedding space. However, CLIP models often develop multimodal spurious biases, which is the undesirable tendency to rely on spurious features. For example, CLIP may infer object types in images based on frequently co-occurring backgrounds rather than the object's core features. This bias significantly impairs the robustness of pre-trained CLIP models on out-of-distribution data, where such cross-modal associations no longer hold. Existing methods for mitigating multimodal spurious bias typically require fine-tuning on downstream data or prior knowledge of the bias, which undermines the out-of-the-box usability of CLIP. In this paper, we first theoretically analyze the impact of multimodal spurious bias in zero-shot classification. Based on this insight, we propose Spuriousness-Aware Guided Exploration (SAGE), a simple and effective method that mitigates spurious bias through guided prompt selection. SAGE requires no training, fine-tuning, or external annotations. It explores a space of prompt templates and selects the prompts that induce the largest semantic separation between classes, thereby improving worst-group robustness. Extensive experiments on four real-world benchmark datasets and five popular backbone models demonstrate that SAGE consistently improves zero-shot performance and generalization, outperforming previous zero-shot approaches without any external knowledge or model updates.