---
title: "HarmTrace：锚定校准解耦优化实现有害模因细粒度目标识别"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.16622v1"
permalink: "https://feed.astralor.com/items/itm_a1aa56da100fb077f52c0c9c"
published: "2026-08-17T14:21:41.000Z"
score: 81
channel: "ai"
tags: ["arXiv","HarmTrace","JRA","Meme3W","多模态大模型","有害内容检测"]
---

# HarmTrace：锚定校准解耦优化实现有害模因细粒度目标识别

## AI 点评

> 这篇论文最值得关注的是把‘检测’从二元分类推向‘问责式识别’，对内容审核从业者来说，能定位攻击目标比单纯打标签有用得多。JRA的提出也让模型短板无处遁形，强烈建议多模态安全方向的研究者阅读。

## 速览

现有有害模因检测常仅判断图文整体是否有害，模型可能猜对有害性却错判攻击目标或支撑证据。为此，论文将检测扩展为细粒度目标识别，要求模型同时输出目标类别、实体、文本提及和视觉区域。作者整合多份公开数据集构建Meme3W，并提出联合记录准确率JRA作为严格评估指标。实验显示主流多模态大模型的有害性准确率与JRA差距悬殊。为弥合差距，提出HarmTrace框架，通过实体感知监督微调强化目标实体监督，并采用条件目标识别策略优化CTPO解耦有害性与目标识别优势，借助虚拟正锚点VPA实现优势归一化。在Qwen3-VL-8B上JRA从17.58%大幅提升至52.51%，代码已开源。

## 关键点

- 提出细粒度目标识别任务，要求模型输出目标类别、实体、文本提及和视觉区域
- 构建Meme3W数据集，统一多份公开有害模因数据并提供人工校验标注
- 提出JRA严格记录级指标，要求有害性标签与所有目标识别字段共同正确
- HarmTrace在Qwen3-VL-8B上JRA从17.58%提升至52.51%

## 全文

Multimodal harmful meme detection is typically formulated as image--text harmfulness classification. A model may correctly predict harmfulness while misidentifying the attacked target or its supporting evidence. We therefore extend harmful meme detection with fine-grained target identification, asking what type of target is attacked, who is targeted, and where the target appears in the meme. The model predicts harmfulness for every meme and, for harmful memes, outputs the target category, target entity, textual mention, and visual region. To support this task, we introduce Meme3W, which unifies multiple public harmful meme datasets and provides human-verified annotations for harmful instances. We further introduce Joint Record Accuracy (JRA), a strict record-level metric requiring the harmfulness label and all target-identification fields to be jointly correct. Experiments with representative multimodal large language models reveal a substantial gap between harmfulness accuracy and JRA. To narrow this gap, we propose HarmTrace, an anchor-calibrated decoupled optimization framework. HarmTrace strengthens target-entity supervision through entity-aware supervised fine-tuning. It then applies Conditional Target-identification Policy Optimization (CTPO) to decouple harmfulness and target-identification advantages, restricting target-identification optimization to label-correct responses for harmful examples. CTPO uses a Virtual Positive Anchor (VPA) as a fully correct reference for target-identification advantage normalization. HarmTrace improves both JRA and harmfulness accuracy across the evaluated backbones, with JRA on the Qwen3-VL-8B backbone increasing from 17.58\% to 52.51\%. Our code is publicly available at https://github.com/llly1234/HarmTrace-for-Harmful-Memes.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_a1aa56da100fb077f52c0c9c