用强化学习生成挑战性问题，暴露视觉语言模型漏洞
FuzzingRL: Reinforcement Fuzz-Testing for Revealing VLM Failures
- 通过视觉与语言模糊生成多样化问题，驱动模型出错
- 使Qwen2.5-VL-32B准确率从86.58%降至65.53%
- 训练好的策略可跨模型复用，适用于多种视觉语言模型
视觉语言模型（VLMs）容易出错，识别其错误发生位置对确保AI系统可靠性与安全性至关重要。本文提出一种自动生成诱导错误响应的问题的方法，以揭示VLM的脆弱性。核心思想是结合模糊测试与对抗性强化微调：通过视觉和语言模糊将单一输入查询生成大量变体，并基于模糊结果，利用对抗性强化微调进一步优化问题生成器，使其产出越来越具挑战性的查询以触发模型失败。该方法可持续降低目标VLM的回答准确率——例如，Qwen2.5-VL-32B在四轮强化学习迭代后，准确率从86.58%降至65.53%。此外，针对单个目标模型训练的模糊策略可迁移至多个其他VLM，生成有效挑战问题并导致其性能下降。
原文摘要 · Abstract (English)
Vision Language Models (VLMs) are prone to errors, and identifying where these errors occur is critical for ensuring the reliability and safety of AI systems. In this paper, we propose an approach that automatically generates questions designed to deliberately induce incorrect responses from VLMs, thereby revealing their vulnerabilities. The core of this approach lies in fuzz testing and reinforcement finetuning: we transform a single input query into a large set of diverse variants through vision and language fuzzing. Based on the fuzzing outcomes, the question generator is further instructed by adversarial reinforcement fine-tuning to produce increasingly challenging queries that trigger model failures. With this approach, we can consistently drive down a target VLM's answer accuracy -- for example, the accuracy of Qwen2.5-VL-32B on our generated questions drops from 86.58\% to 65.53\% in four RL iterations. Moreover, a fuzzing policy trained against a single target VLM transfers to multiple other VLMs, producing challenging queries that degrade their performance as well.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。