SC-Captioner: Improving Image Captioning with Self-Correction by Reinforcement Learning

• URL: http://arxiv.org/abs/2508.06125v1
• Date: Fri, 08 Aug 2025 08:45:52 GMT
• Title: SC-Captioner: Improving Image Captioning with Self-Correction by Reinforcement Learning
• Authors: Lin Zhang, Xianfang Zeng, Kangcong Li, Gang Yu, Tao Chen,
• Abstract summary: SC-Captioner is a reinforcement learning framework that enables the self-correcting capability of image caption models.<n>We calculate the set difference between sets of initial and self-corrected captions to identify added and removed elements.<n>Experiments show that applying SC-Captioner on large visual-language models can generate better image captions across various scenarios.
• Score: 21.739084696595427
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose SC-Captioner, a reinforcement learning framework that enables the self-correcting capability of image caption models. Our crucial technique lies in the design of the reward function to incentivize accurate caption corrections. Specifically, the predicted and reference captions are decomposed into object, attribute, and relation sets using scene-graph parsing algorithms. We calculate the set difference between sets of initial and self-corrected captions to identify added and removed elements. These elements are matched against the reference sets to calculate correctness bonuses for accurate refinements and mistake punishments for wrong additions and removals, thereby forming the final reward. For image caption quality assessment, we propose a set of metrics refined from CAPTURE that alleviate its incomplete precision evaluation and inefficient relation matching problems. Furthermore, we collect a fine-grained annotated image caption dataset, RefinedCaps, consisting of 6.5K diverse images from COCO dataset. Experiments show that applying SC-Captioner on large visual-language models can generate better image captions across various scenarios, significantly outperforming the direct preference optimization training strategy.

Related papers

• CCCaption: Dual-Reward Reinforcement Learning for Complete and Correct Image Captioning [23.289413412387223]
  We introduce CCCaption: a dual-reward reinforcement learning framework with a dedicated fine-tuning corpus.<n>For completeness, we use diverse LVLMs to disentangle the image into a set of visual queries, and reward captions that answer more of these queries.<n>For correctness, we penalize captions that contain hallucinations by validating the authenticity of sub-caption queries.
  arXiv  Detail & Related papers  (2026-02-25T07:34:26Z)
• ScaleCap: Inference-Time Scalable Image Captioning via Dual-Modality Debiasing [128.8346376825612]
  Key challenges of high-quality image captioning lie in the inherent biases of LVLMs.<n>We propose a scalable debiased captioning strategy, which continuously enriches and calibrates the caption with increased inference budget.<n>Annotating 450K images with ScaleCap and using them for LVLM pretraining leads to consistent performance gains across 11 widely used benchmarks.
  arXiv  Detail & Related papers  (2025-06-24T17:59:55Z)
• Fluent and Accurate Image Captioning with a Self-Trained Reward Model [47.213906345208315]
  We propose Self-Cap, a captioning approach that relies on a learnable reward model based on self-generated negatives. Our discriminator is a fine-tuned contrastive image-text model trained to promote caption correctness.
  arXiv  Detail & Related papers  (2024-08-29T18:00:03Z)
• ASPIRE: Language-Guided Data Augmentation for Improving Robustness Against Spurious Correlations [43.323791505213634]
  ASPIRE (Language-guided Data Augmentation for SPurIous correlation REmoval) is a solution for supplementing the training dataset with images without spurious features. It can generate non-spurious images without requiring any group labeling or existing non-spurious images in the training set. It improves the worst-group classification accuracy of prior methods by 1% - 38%.
  arXiv  Detail & Related papers  (2023-08-19T20:18:15Z)
• ALIP: Adaptive Language-Image Pre-training with Synthetic Caption [78.93535202851278]
  Contrastive Language-Image Pre-training (CLIP) has significantly boosted the performance of various vision-language tasks. The presence of intrinsic noise and unmatched image-text pairs in web data can potentially affect the performance of representation learning. We propose an Adaptive Language-Image Pre-training (ALIP), a bi-path model that integrates supervision from both raw text and synthetic caption.
  arXiv  Detail & Related papers  (2023-08-16T15:19:52Z)
• Simple Token-Level Confidence Improves Caption Correctness [117.33497608933169]
  Token-Level Confidence, or TLC, is a simple yet surprisingly effective method to assess caption correctness. We fine-tune a vision-language model on image captioning, input an image and proposed caption to the model, and aggregate token confidences over words or sequences to estimate image-caption consistency.
  arXiv  Detail & Related papers  (2023-05-11T17:58:17Z)
• Large-Scale Bidirectional Training for Zero-Shot Image Captioning [44.17587735943739]
  We introduce Bidirectional Image Text Training in largER Scale, BITTERS, an efficient training and inference framework for zero-shot image captioning. We show that careful selection of large-scale training set and model architecture is the key to achieving zero-shot image captioning.
  arXiv  Detail & Related papers  (2022-11-13T00:09:36Z)
• Prompt-based Learning for Unpaired Image Captioning [86.44188293709307]
  Unpaired Image Captioning (UIC) has been developed to learn image descriptions from unaligned vision-language sample pairs. Recent successes of Vision-Language Pre-Trained Models (VL-PTMs) have triggered the development of prompt-based learning. We present in this paper a novel scheme based on prompt to train the UIC model, making best use of the powerful generalization ability.
  arXiv  Detail & Related papers  (2022-05-26T03:13:43Z)
• Fine-grained Image Captioning with CLIP Reward [104.71533106301598]
  We propose using CLIP, a multimodal encoder trained on huge image-text pairs from web, to calculate multimodal similarity and use it as a reward function. We also propose a simple finetuning strategy of the CLIP text encoder to improve grammar that does not require extra text annotation. In experiments on text-to-image retrieval and FineCapEval, the proposed CLIP-guided model generates more distinctive captions than the CIDEr-optimized model.
  arXiv  Detail & Related papers  (2022-05-26T02:46:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.