---
title: "BanglaWild：面向孟加拉语自然场景文本识别的OCR与视觉语言模型基准"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.03884v1"
permalink: "https://feed.astralor.com/items/itm_e927b258a7751066547f2819"
published: "2026-08-04T16:20:53.000Z"
score: 77
channel: "ai"
tags: ["arXiv","BANGLAWILD","LoRA","场景文本识别","多语种OCR"]
---

# BanglaWild：面向孟加拉语自然场景文本识别的OCR与视觉语言模型基准

## AI 点评

> 这篇论文最值得关注的是它把传统OCR和VLM拉到同一张考卷上，然后用细粒度错误分类打破了一个老生常谈的假设——在孟加拉语里连字问题根本不是主要瓶颈，视觉误识别才是。如果你想做低资源语言场景文本识别，或者想评估自家VLM的细粒度能力，这个基准和错误分类体系非常值得借鉴。

## 速览

孟加拉语自然场景文本识别长期缺乏统一评测：现有资源多针对手写文档或受限标牌，指标单一且未同时覆盖OCR与视觉语言模型。本论文推出BANGLAWILD基准，包含2,535张孟加拉语场景文本图像，配有逐字标注、分类轴、诊断属性和拼写规范形式。研究评估了15个视觉语言模型和3个传统OCR系统，结合多种提示策略、LoRA微调与LLM-as-a-Judge评测，发现同系列大模型未必优于小模型，视觉错误占主导约60%，辅音连字错误不足2%，挑战了既有假设。

## 关键点

- BANGLAWILD包含2,535张孟加拉语场景文本图像，带逐字标注、两个分类轴、四个诊断属性和拼写规范形式。
- 评估了15个视觉语言模型和3个传统OCR系统，使用三种提示策略并对6个开源模型进行LoRA微调。
- 发现同系列中更大模型并未超越更小模型；视觉误识别约占总错误60%，辅音连字错误不足2%。
- 提示语言主要影响跨脚本漂移；LoRA可减少弱模型的严重失败，但不提升强模型上限。

## 全文

In-the-wild Bengali scene text recognition is largely unmeasured: existing resources target handwritten documents or constrained sign-board parsing, report only aggregate edit-distance metrics, and evaluate either conventional OCR or VLMs, never both on the same in-the-wild data. To address this gap, we introduce BANGLAWILD, a benchmark of 2,535 Bengali scene text images, each paired with a verbatim gold transcription, two categorical axes, four diagnostic attributes, and an orthographically standard form where the in-image text deviates from canonical spelling. We evaluate fifteen VLMs and three conventional OCR systems under three prompting strategies, fine-tune 6 open-source models with LoRA, and complement edit-distance metrics with an LLM-as-a-Judge evaluation. Our results reveal a persistent gap in which larger models within the same family do not outperform smaller ones. Our fifteen-class error taxonomy shows that visual mis-recognition accounts for ~60% of errors in the strongest systems, while conjunct-related errors contribute under 2%, challenging a long-standing assumption in Bengali OCR research; the same visual dominant profile also holds across architectures, including the one conventional baseline that reads Bengali reliably. Prompt language mainly affects cross-script drift and LoRA reduces catastrophic failures in weak models without lifting the ceiling on already competent ones. Code and data will be publicly released.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_e927b258a7751066547f2819