ARC-Hunyuan-Video-7B: Structured Video Comprehension of Real-World Shorts

TL;DR

ARC-Hunyuan-Video-7B is a multimodal model designed for structured comprehension of real-world short videos, enabling detailed understanding, captioning, question answering, and grounding, with strong performance and efficiency.

Contribution

The paper introduces a novel 7B-parameter multimodal model capable of end-to-end structured video comprehension for complex real-world shorts, with a comprehensive training regimen and new benchmark.

Findings

Strong performance on ShortVid-Bench benchmark

Effective zero-shot and few-shot downstream application support

Real-world deployment improves user engagement and satisfaction

Abstract

Real-world user-generated short videos, especially those distributed on platforms such as WeChat Channel and TikTok, dominate the mobile internet. However, current large multimodal models lack essential temporally-structured, detailed, and in-depth video comprehension capabilities, which are the cornerstone of effective video search and recommendation, as well as emerging video applications. Understanding real-world shorts is actually challenging due to their complex visual elements, high information density in both visuals and audio, and fast pacing that focuses on emotional expression and viewpoint delivery. This requires advanced reasoning to effectively integrate multimodal information, including visual, audio, and text. In this work, we introduce ARC-Hunyuan-Video, a multimodal model that processes visual, audio, and textual signals from raw video inputs end-to-end for structured comprehension. The model is capable of multi-granularity timestamped video captioning and summarization, open-ended video question answering, temporal video grounding, and video reasoning. Leveraging high-quality data from an automated annotation pipeline, our compact 7B-parameter model is trained through a comprehensive regimen: pre-training, instruction fine-tuning, cold start, reinforcement learning (RL) post-training, and final instruction fine-tuning. Quantitative evaluations on our introduced benchmark ShortVid-Bench and qualitative comparisons demonstrate its strong performance in real-world video comprehension, and it supports zero-shot or fine-tuning with a few samples for diverse downstream applications. The real-world production deployment of our model has yielded tangible and measurable improvements in user engagement and satisfaction, a success supported by its remarkable efficiency, with stress tests indicating an inference time of just 10 seconds for a one-minute video on H20 GPU.