RoadTones: Tone Controllable Text Generation from Road Event Videos

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Parikh, Chirag, Lipare, Siddhi Pravin, Sarvadevabhatla, Ravi Kiran
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918514701369344
author Parikh, Chirag
Lipare, Siddhi Pravin
Sarvadevabhatla, Ravi Kiran
author_facet Parikh, Chirag
Lipare, Siddhi Pravin
Sarvadevabhatla, Ravi Kiran
contents Existing video-language models can generate factual descriptions of road events but lack control over how these events are expressed: their tone, urgency, or style. This limits deployment in communication-critical settings where the effectiveness of a message depends on both content and presentation, not just factual accuracy. To mitigate this, we introduce a comprehensive dataset-model-evaluation suite for tone-controllable road video captioning. Our human-validated data generation pipeline expands road-video corpora with diverse tonal annotations and multi-tone captions, yielding the RoadTones-51K dataset. We propose RoadTones-VL-CoT, a controllable video-to-text model that also generates tone-conditioned Chain-of-Thought intermediate drafts for interpretability. We also introduce RoadTones-Eval, a new evaluation suite that jointly measures factual consistency and tone adherence. In addition, we conducted a user study whose results validate caption quality, tone control, and factual consistency. Together, these contributions lay the foundation for context-sensitive tone-controllable video captioning.
format Preprint
id arxiv_https___arxiv_org_abs_2605_21411
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RoadTones: Tone Controllable Text Generation from Road Event Videos
Parikh, Chirag
Lipare, Siddhi Pravin
Sarvadevabhatla, Ravi Kiran
Computer Vision and Pattern Recognition
Existing video-language models can generate factual descriptions of road events but lack control over how these events are expressed: their tone, urgency, or style. This limits deployment in communication-critical settings where the effectiveness of a message depends on both content and presentation, not just factual accuracy. To mitigate this, we introduce a comprehensive dataset-model-evaluation suite for tone-controllable road video captioning. Our human-validated data generation pipeline expands road-video corpora with diverse tonal annotations and multi-tone captions, yielding the RoadTones-51K dataset. We propose RoadTones-VL-CoT, a controllable video-to-text model that also generates tone-conditioned Chain-of-Thought intermediate drafts for interpretability. We also introduce RoadTones-Eval, a new evaluation suite that jointly measures factual consistency and tone adherence. In addition, we conducted a user study whose results validate caption quality, tone control, and factual consistency. Together, these contributions lay the foundation for context-sensitive tone-controllable video captioning.
title RoadTones: Tone Controllable Text Generation from Road Event Videos
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.21411