SOURCE-LINKED INTELLIGENCE

Learning to Zoom Efficiently with a Contrastive Curriculum

arXiv · AI, language, vision and robotics · article · Sep 2, 2026 · UTC

Using a zoom-in tool is an important foundational part of modern visual agents, because it allows to efficiently handle tasks involving high-resolution images. Most previous methods need an extensive warm-start supervised fine-tuning phase for teaching models zoom-in. We show that this is not necessary by proposing a new intrinsic reward for learning tool use in MLLMs without the need for additional labels or warm-start SFT. Our InfoNCE-style reward uses a curriculum of increasingly hard negative tool calls as a contrastive training signal. Empirical experiments on $V^*$, HRBench and MME-RealW

Read original source ↗ Open in workspace

recordType

paper

region

Global