視覺對比式自蒸餾

摘要

同策略自蒸馏（OPSD）因其无需同策略蒸馏（OPD）所需的外部教师模型而具有前景，但其仍需教师与学生之间存在信息不对称，以确保自教师提供比学生更强的学习信号。现有方法通过特权答案或视觉证据来创造这种不对称性。我们探究是否可同时消除这两者，从而得到一种纯粹由输入条件驱动、形式更简化的OPSD。为此，我们提出视觉对比自蒸馏方法VCSD，将图像内容移除转化为同策略自蒸馏信号。在每个学生生成的响应前缀处，EMA教师在同一提示与前缀下产生两个下一词元分布——一个以原始图像为条件，另一个以内容擦除后的控制图为条件。两者在词元级别的对数概率差异，突出显示了那些因实例级视觉内容而可能性显著提升的候选词元。我们利用这种对比，在教师原始图像分布的合理支持范围内锐化该分布，并将所得全分布目标蒸馏至学生模型。基于ViRL39K数据集，VCSD在Qwen3-VL与Qwen3.5模型上始终优于对照的OPSD方法。例如，在Qwen3-VL上，VCSD将七个基准的聚合得分从2B规模的62.27%提升至67.04%，4B规模从71.30%提升至73.16%，8B规模从72.51%提升至76.26%。此外，VCSD无需外部教师、特权答案、视觉证据信号、推理轨迹，也不增加额外的推理时开销。

English

On-policy self-distillation (OPSD) is promising as it removes the external teacher required by on-policy distillation (OPD), yet it still needs asymmetric information between teacher and student to ensure that the self-teacher provides a stronger learning signal than the student. Existing methods create this asymmetry either through privileged answers or visual evidence. We ask whether both can be removed, yielding a simpler form of OPSD driven purely by input conditioning. For this purpose, we propose Visual Contrastive Self-Distillation, namely VCSD, which converts image-content removal into an on-policy self-distillation signal. At each student-generated response prefix, the EMA teacher produces two next-token distributions under the same prompt and prefix -- one conditioned on the original image and the other on a content-erased control. Their token-wise log-probability difference highlights candidates whose likelihood is specifically increased by the instance-level visual content. We use this contrast to sharpen the teacher's original-image distribution within its plausible support, and distill the resulting full-distribution target into the student. Using ViRL39K dataset, VCSD consistently outperforms matched OPSD across Qwen3-VL and Qwen3.5 models. For example, on Qwen3-VL, it improves the seven-benchmark aggregate from 62.27% rightarrow 67.04% at 2B, 71.30% rightarrow 73.16% at 4B, and 72.51% rightarrow 76.26% at 8B. Furthermore, VCSD requires no external teacher, privileged answers, visual evidence signals, reasoning traces, or additional inference-time cost.