CoFi-Dec: Hallucination-Resistant Decoding via Coarse-to-Fine Generative Feedback in Large Vision-Language Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917173930229760 |
|---|---|
| author | Cao, Zongsheng He, Yangfan Liu, Anran Xie, Jun Chen, Feng Wang, Zepeng |
| author_facet | Cao, Zongsheng He, Yangfan Liu, Anran Xie, Jun Chen, Feng Wang, Zepeng |
| contents | Large Vision-Language Models (LVLMs) have achieved impressive progress in multi-modal understanding and generation. However, they still tend to produce hallucinated content that is inconsistent with the visual input, which limits their reliability in real-world applications. We propose \textbf{CoFi-Dec}, a training-free decoding framework that mitigates hallucinations by integrating generative self-feedback with coarse-to-fine visual conditioning. Inspired by the human visual process from global scene perception to detailed inspection, CoFi-Dec first generates two intermediate textual responses conditioned on coarse- and fine-grained views of the original image. These responses are then transformed into synthetic images using a text-to-image model, forming multi-level visual hypotheses that enrich grounding cues. To unify the predictions from these multiple visual conditions, we introduce a Wasserstein-based fusion mechanism that aligns their predictive distributions into a geometrically consistent decoding trajectory. This principled fusion reconciles high-level semantic consistency with fine-grained visual grounding, leading to more robust and faithful outputs. Extensive experiments on six hallucination-focused benchmarks show that CoFi-Dec substantially reduces both entity-level and semantic-level hallucinations, outperforming existing decoding strategies. The framework is model-agnostic, requires no additional training, and can be seamlessly applied to a wide range of LVLMs. The implementation is available at https://github.com/AI-Researcher-Team/CoFi-Dec. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_23453 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CoFi-Dec: Hallucination-Resistant Decoding via Coarse-to-Fine Generative Feedback in Large Vision-Language Models Cao, Zongsheng He, Yangfan Liu, Anran Xie, Jun Chen, Feng Wang, Zepeng Computer Vision and Pattern Recognition Artificial Intelligence Large Vision-Language Models (LVLMs) have achieved impressive progress in multi-modal understanding and generation. However, they still tend to produce hallucinated content that is inconsistent with the visual input, which limits their reliability in real-world applications. We propose \textbf{CoFi-Dec}, a training-free decoding framework that mitigates hallucinations by integrating generative self-feedback with coarse-to-fine visual conditioning. Inspired by the human visual process from global scene perception to detailed inspection, CoFi-Dec first generates two intermediate textual responses conditioned on coarse- and fine-grained views of the original image. These responses are then transformed into synthetic images using a text-to-image model, forming multi-level visual hypotheses that enrich grounding cues. To unify the predictions from these multiple visual conditions, we introduce a Wasserstein-based fusion mechanism that aligns their predictive distributions into a geometrically consistent decoding trajectory. This principled fusion reconciles high-level semantic consistency with fine-grained visual grounding, leading to more robust and faithful outputs. Extensive experiments on six hallucination-focused benchmarks show that CoFi-Dec substantially reduces both entity-level and semantic-level hallucinations, outperforming existing decoding strategies. The framework is model-agnostic, requires no additional training, and can be seamlessly applied to a wide range of LVLMs. The implementation is available at https://github.com/AI-Researcher-Team/CoFi-Dec. |
| title | CoFi-Dec: Hallucination-Resistant Decoding via Coarse-to-Fine Generative Feedback in Large Vision-Language Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2512.23453 |