CVAINov 9, 2025

Zooming into Comics: Region-Aware RL Improves Fine-Grained Comic Understanding in Vision-Language Models

arXiv:2511.06490v1h-index: 7Has Code
Originality Incremental advance
AI Analysis

This addresses the problem of VLMs performing poorly on comics for researchers and developers, but it is incremental as it builds on existing methods like RL and fine-tuning.

The paper tackled the challenge of Vision-Language Models (VLMs) struggling with fine-grained comic understanding by introducing the AI4VA-FG benchmark and proposing Region-Aware Reinforcement Learning (RARL), which improved entity recognition and storyline ordering in the Qwen2.5-VL model.

Complex visual narratives, such as comics, present a significant challenge to Vision-Language Models (VLMs). Despite excelling on natural images, VLMs often struggle with stylized line art, onomatopoeia, and densely packed multi-panel layouts. To address this gap, we introduce AI4VA-FG, the first fine-grained and comprehensive benchmark for VLM-based comic understanding. It spans tasks from foundational recognition and detection to high-level character reasoning and narrative construction, supported by dense annotations for characters, poses, and depth. Beyond that, we evaluate state-of-the-art proprietary models, including GPT-4o and Gemini-2.5, and open-source models such as Qwen2.5-VL, revealing substantial performance deficits across core tasks of our benchmarks and underscoring that comic understanding remains an unsolved challenge. To enhance VLMs' capabilities in this domain, we systematically investigate post-training strategies, including supervised fine-tuning on solutions (SFT-S), supervised fine-tuning on reasoning trajectories (SFT-R), and reinforcement learning (RL). Beyond that, inspired by the emerging "Thinking with Images" paradigm, we propose Region-Aware Reinforcement Learning (RARL) for VLMs, which trains models to dynamically attend to relevant regions through zoom-in operations. We observe that when applied to the Qwen2.5-VL model, RL and RARL yield significant gains in low-level entity recognition and high-level storyline ordering, paving the way for more accurate and efficient VLM applications in the comics domain.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes