mlsanityData to 7 Oct 2026

A Baseline for Detecting Out-of-Distribution Examples in Image Captioning

arXiv 2207.05418 ↗First posted 12 Jul 2022cs.CV cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

Image captioning research achieved breakthroughs in recent years by developing neural models that can generate diverse and high-quality descriptions for images drawn from the same distribution as training images. However, when facing out-of-distribution (OOD) images, such as corrupted images, or images containing unknown objects, the models fail in generating relevant captions. In this paper, we consider the problem of OOD detection in image captioning. We formulate the problem and suggest an evaluation setup for assessing the model's performance on the task. Then, we analyze and show the effectiveness of the caption's likelihood score at detecting and rejecting OOD images, which implies that the relatedness between the input image and the generated caption is encapsulated within the score.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 29 of its 44 references that are in the map. First posted on arXiv 12 Jul 2022.

Builds on

29 of 44 references.

References · 29

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  2. 2017On Calibration of Modern Neural Networkscited by 4,860
  3. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  4. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  5. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
  6. 2014CIDEr: Consensus-based Image Description Evaluationcited by 2,027
  7. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,472
  8. 2017Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networkscited by 1,325
  9. 2017Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answeringcited by 1,310
  10. 2018Hierarchical Neural Story Generationcited by 1,182
  11. 2014Show and Tell: A Neural Image Caption Generatorcited by 877
  12. 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 774
17 more
  1. 2018nocaps: novel object captioning at scalecited by 444
  2. 2020Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Datacited by 408
  3. 2017Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samplescited by 389
  4. 2019Meshed-Memory Transformer for Image Captioningcited by 343
  5. 2020Pretrained Transformers Improve Out-of-Distribution Robustnesscited by 293
  6. 2019Image Captioning: Transforming Objects into Wordscited by 171
  7. 2018Neural Baby Talkcited by 107
  8. 2020Improving Image Captioning by Leveraging Intra- and Inter-layer Global Representation in Transformer Networkcited by 48
  9. 2021OODformer: Out-Of-Distribution Detection Transformercited by 27
  10. 2018Out-of-Distribution Detection using Multiple Semantic Label Representationscited by 22
  11. 2020Image Captioning as an Assistive Technology: Lessons Learned from VizWiz 2020 Challengecited by 17
  12. 2016Rich Image Captioning in the Wildcited by 11
  13. 2020Self-Supervised Out-of-Distribution Detection in Brain CT Scanscited by 10
  14. 2018Actigraphy-based Sleep/Wake Pattern Detection using Convolutional Neural Networkscited by 2
  15. 2019Generating Diverse and Informative Natural Language Fashion Feedbackcited by 2
  16. 2019Joint Visual-Textual Embedding for Multimodal Style Searchcited by 2
  17. 2020Redesigning the classification layer by randomizing the class representation vectorscited by 2

Built on by

Cited by year

Citing papers per year: 2024: 1; 2025: 3.

4 papers.

  1. 2024Uncertainty quantification for improving radiomic-based models in radiation pneumonitis predictioncited by 0
  2. 2025Disentangling Fine-Tuning from Pre-Training in Visual Captioning with Hybrid Markov Logiccited by 0
  3. 2025On Explaining Visual Captioning with Hybrid Markov Logic Networkscited by 0
  4. 2025Simplex-Optimized Hybrid Ensemble for Large Language Model Text Detection Under Generative Distribution Drifcited by 0