# See or Guess: Counterfactually Regularized Image Captioning

arXiv: https://arxiv.org/abs/2408.16809 · HTML: https://mlsanity.com/paper/2408.16809/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 50 references that are in the map.
- First posted on arXiv 29 Aug 2024.

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

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models](https://mlsanity.com/paper/2304.10592/) (2023)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [Show, Attend and Tell: Neural Image Caption Generation with Visual Attention](https://mlsanity.com/paper/1502.03044/) (2015)
- [Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models](https://mlsanity.com/paper/1505.04870/) (2015)
- [Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks](https://mlsanity.com/paper/2004.06165/) (2020)
- [Show and Tell: A Neural Image Caption Generator](https://mlsanity.com/paper/1411.4555/) (2014)
- [ClipCap: CLIP Prefix for Image Captioning](https://mlsanity.com/paper/2111.09734/) (2021)
- [Unified Vision-Language Pre-Training for Image Captioning and VQA](https://mlsanity.com/paper/1909.11059/) (2019)
- [Counterfactual Visual Explanations](https://mlsanity.com/paper/1904.07451/) (2019)
- [Direct and Indirect Effects](https://mlsanity.com/paper/1301.2300/) (2013)
- [Counterfactual VQA: A Cause-Effect Look at Language Bias](https://mlsanity.com/paper/2006.04315/) (2020)
- [From Show to Tell: A Survey on Deep Learning-based Image Captioning](https://mlsanity.com/paper/2107.06912/) (2021)
- [CXPlain: Causal Explanations for Model Interpretation under Uncertainty](https://mlsanity.com/paper/1910.12336/) (2019)
- [Let there be a clock on the beach: Reducing Object Hallucination in Image Captioning](https://mlsanity.com/paper/2110.01705/) (2021)
- [Plausible May Not Be Faithful: Probing Object Hallucination in Vision-Language Pre-training](https://mlsanity.com/paper/2210.07688/) (2022)
- [Learning What Makes a Difference from Counterfactual Examples and Gradient Supervision](https://mlsanity.com/paper/2004.09034/) (2020)
- [Deconfounded Image Captioning: A Causal Retrospect](https://mlsanity.com/paper/2003.03923/) (2020)
- [COCO-Counterfactuals: Automatically Constructed Counterfactual Examples for Image-Text Pairs](https://mlsanity.com/paper/2309.14356/) (2023)
- [Generating Counterfactual Explanations with Natural Language](https://mlsanity.com/paper/1806.09809/) (2018)
- [Consensus Graph Representation Learning for Better Grounded Image Captioning](https://mlsanity.com/paper/2112.00974/) (2021)
- [With a Little Help from your own Past: Prototypical Memory Networks for Image Captioning](https://mlsanity.com/paper/2308.12383/) (2023)
- [Counterfactuals and Policy Analysis in Structural Models](https://mlsanity.com/paper/1302.4929/) (2013)
- [Counterfactual Maximum Likelihood Estimation for Training Deep Networks](https://mlsanity.com/paper/2106.03831/) (2021)
- [Multi-Modal Experience Inspired AI Creation](https://mlsanity.com/paper/2209.02427/) (2022)

## Abstract (arXiv)

Image captioning, which generates natural language descriptions of the visual information in an image, is a crucial task in vision-language research. Previous models have typically addressed this task by aligning the generative capabilities of machines with human intelligence through statistical fitting of existing datasets. While effective for normal images, they may struggle to accurately describe those where certain parts of the image are obscured or edited, unlike humans who excel in such cases. These weaknesses they exhibit, including hallucinations and limited interpretability, often hinder performance in scenarios with shifted association patterns. In this paper, we present a generic image captioning framework that employs causal inference to make existing models more capable of interventional tasks, and counterfactually explainable. Our approach includes two variants leveraging either total effect or natural direct effect. Integrating them into the training process enables models to handle counterfactual scenarios, increasing their generalizability. Extensive experiments on various datasets show that our method effectively reduces hallucinations and improves the model's faithfulness to images, demonstrating high portability across both small-scale and large-scale image-to-text models. The code is available at https://github.com/Aman-4-Real/See-or-Guess.