RONA: Pragmatically Diverse Image Captioning with Coherence Relations
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916785447501824 |
|---|---|
| author | Ramakrishnan, Aashish Anantha Ramakrishnan, Aadarsh Anantha Lee, Dongwon |
| author_facet | Ramakrishnan, Aashish Anantha Ramakrishnan, Aadarsh Anantha Lee, Dongwon |
| contents | Writing Assistants (e.g., Grammarly, Microsoft Copilot) traditionally generate diverse image captions by employing syntactic and semantic variations to describe image components. However, human-written captions prioritize conveying a central message alongside visual descriptions using pragmatic cues. To enhance caption diversity, it is essential to explore alternative ways of communicating these messages in conjunction with visual content. We propose RONA, a novel prompting strategy for Multi-modal Large Language Models (MLLM) that leverages Coherence Relations as a controllable axis for pragmatic variations. We demonstrate that RONA generates captions with better overall diversity and ground-truth alignment, compared to MLLM baselines across multiple domains. Our code is available at: https://github.com/aashish2000/RONA |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_10997 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RONA: Pragmatically Diverse Image Captioning with Coherence Relations Ramakrishnan, Aashish Anantha Ramakrishnan, Aadarsh Anantha Lee, Dongwon Computation and Language Artificial Intelligence Computer Vision and Pattern Recognition 68T50 I.2.7; I.2.10 Writing Assistants (e.g., Grammarly, Microsoft Copilot) traditionally generate diverse image captions by employing syntactic and semantic variations to describe image components. However, human-written captions prioritize conveying a central message alongside visual descriptions using pragmatic cues. To enhance caption diversity, it is essential to explore alternative ways of communicating these messages in conjunction with visual content. We propose RONA, a novel prompting strategy for Multi-modal Large Language Models (MLLM) that leverages Coherence Relations as a controllable axis for pragmatic variations. We demonstrate that RONA generates captions with better overall diversity and ground-truth alignment, compared to MLLM baselines across multiple domains. Our code is available at: https://github.com/aashish2000/RONA |
| title | RONA: Pragmatically Diverse Image Captioning with Coherence Relations |
| topic | Computation and Language Artificial Intelligence Computer Vision and Pattern Recognition 68T50 I.2.7; I.2.10 |
| url | https://arxiv.org/abs/2503.10997 |