减少自动驾驶VLMs中合成生成MCQA的文本偏差
Reducing Text Bias in Synthetically Generated MCQAs for VLMs in Autonomous Driving
AI总结：
本文提出方法减少自动驾驶VLMs中合成生成MCQA的文本偏差，通过解耦正确答案与语言痕迹并采用课程学习策略，使模型依赖视觉基础以提升感知理解能力。
摘要：
多选题回答（MCQA）基准是衡量视觉语言模型（VLM）在驾驶任务中性能的标准。然而，我们观察到合成生成的MCQA对隐藏的文本线索高度敏感，使模型能够利用语言模式而非视觉上下文。我们的结果表明，经过此类数据微调的VLM在没有视觉输入的情况下可以达到与人工验证基准相当的准确性。我们提出的方法将盲准确率从随机以上+66.9%降低到+2.9%，消除了绝大多数可利用的文本捷径。通过将正确答案与语言痕迹解耦，并采用课程学习策略，我们迫使模型依赖视觉基础，确保性能准确反映感知理解。
英文摘要：
Multiple Choice Question Answering (MCQA) benchmarks are an established standard for measuring Vision Language Model (VLM) performance in driving tasks. However, we observe the known phenomenon that synthetically generated MCQAs are highly susceptible to hidden textual cues that allow models to exploit linguistic patterns rather than visual context. Our results show that a VLM fine-tuned on such data can achieve accuracy comparable to human-validated benchmarks even without visual input. Our proposed method reduces blind accuracy from +66.9% above random to +2.9%, eliminating the vast majority of exploitable textual shortcuts. By decoupling the correct answer from linguistic artifacts and employing a curriculum learning strategy, we force the model to rely on visual grounding, ensuring that performance accurately reflects perceptual understanding.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
