做AI公平性研究或模型对齐的团队，这篇论文直接戳破了VLM在模糊输入下的性别偏见黑箱——LALS方法让你能逐层看到模型内部编码与输出的脱耦，建议做模型审计的开发者点开看看具体实验设计。
研究发现，视觉语言模型（VLM）在处理性别模糊的图像（如全副武装的工人、背影）时，即使内部编码了女性关联，输出仍倾向于男性，尤其在传统女性职业上表现明显。研究者提出零样本指标LALS，通过将视觉token激活投影到文本嵌入空间，逐层测量概念关联。实验覆盖15个职业、800多张模糊图像和4个VLM，发现模型内部存在不对称过滤：男性信号从头到尾增强，女性信号在中间层达到峰值后被压制。服装颜色等文化线索会进一步调节内部关联。这项研究揭示了VLM在模糊输入下的性别偏见机制，对AI公平性评估有重要启示。
Vision-Language Models Suppress Female Representations Under Ambiguous Input
Alignment teaches vision-language models (VLMs) to avoid expressing demographic biases, and when gender is clearly visible they largely succeed. Far less is known about ambiguous inputs (a worker in full gear, a figure seen from behind) cases common in practice yet rarely studied. We find that minimal prompting pressure exposes occupation-gender defaults when prompting ambiguous input images, with models collapsing to male even for strongly female-stereotyped occupations. But do these outputs reflect what models actually encode internally? We introduce LALS (Latent Association Leaning Score), a zero-shot metric that projects visual-token activations into the model's text-embedding space to measure concept associations per token and layer. Across 15 occupations, over 800 gender-ambiguous images, and four VLMs, internal representations and outputs are systematically decoupled: models often encode a female association internally yet output male. Layer-wise analysis reveals an asymmetric filter -- male signal amplifies end-to-end while female signal peaks mid-network and is suppressed before generation -- and a color ablation shows that culturally loaded visual cues such as clothing color further modulate these internal associations.