FocalOrder: Focal Preference Optimization for Reading Order Detection

TL;DR

FocalOrder introduces a novel training framework that dynamically identifies and emphasizes difficult reading order transitions in documents, significantly improving understanding accuracy over existing methods.

Contribution

The paper proposes FocalOrder with Focal Preference Optimization, addressing positional disparity by focusing on hard-to-learn layout transitions, advancing document reading order detection.

Findings

Achieves state-of-the-art results on OmniDocBench v1.0 and Comp-HRDoc.

Outperforms both specialized baselines and large-scale general VLMs.

Demonstrates the importance of difficulty-aware optimization for complex document structures.

Abstract

Reading order detection is the foundation of document understanding. Most existing methods rely on uniform supervision, implicitly assuming a constant difficulty distribution across layout regions. In this work, we challenge this assumption by revealing a critical flaw: \textbf{Positional Disparity}, a phenomenon where models demonstrate mastery over the deterministic start and end regions but suffer a performance collapse in the complex intermediate sections. This degradation arises because standard training allows the massive volume of easy patterns to drown out the learning signals from difficult layouts. To address this, we propose \textbf{FocalOrder}, a framework driven by \textbf{Focal Preference Optimization (FPO)}. Specifically, FocalOrder employs adaptive difficulty discovery with exponential moving average mechanism to dynamically pinpoint hard-to-learn transitions, while introducing a difficulty-calibrated pairwise ranking objective to enforce global logical consistency. Extensive experiments demonstrate that FocalOrder establishes new state-of-the-art results on OmniDocBench v1.0 and Comp-HRDoc. Our compact model not only outperforms competitive specialized baselines but also significantly surpasses large-scale general VLMs. These results demonstrate that aligning the optimization with intrinsic structural ambiguity of documents is critical for mastering complex document structures.