# 无需强假设：通过时序差异进行视觉表征学习

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-06-14 08:00
- AIHOT 分数：49
- AIHOT 链接：https://aihot.virxact.com/items/cmqguyr0h00t4slpu1yrwqbo4
- 原文链接：https://arxiv.org/abs/2606.15956

## AI 摘要

TDV（Temporal Difference in Vision）是一种新的自监督视觉表征学习范式，放弃数据增强、掩码等强归纳偏置，依赖“过去导致未来”的因果假设。它联合训练图像编码器和运动编码器，使当前帧表示加上编码的运动等于下一帧表示。实验表明，归纳偏置的最优强度随数据量增长而下降。在无需强假设下，TDV在密集空间任务上匹配当前最优方法，为弱假设表征学习奠定基础。

## 正文

Progress in AI has largely been driven by methods that assume less. As compute and data increase, approaches with weaker inductive biases generally outperform those with stronger assumptions. This is particularly characteristic of the field of Visual Representation Learning, where approaches have gone from being dominated by Supervised Learning, to Weakly Supervised Learning, to the now widespread success of Self-Supervised Learning without human labels. Yet, even modern Self-Supervised Learning approaches still depend on strong inductive biases such as augmentations, masking, or cropping. If this trend holds, even these remaining biases should become bottlenecks at scale -- and our experiments confirm this: the optimal strength of inductive biases decreases as data grows. This motivates the search for approaches that rely on fewer assumptions. To this end, we introduce Temporal Difference in Vision (TDV), a new paradigm for self-supervised learning from video that avoids existing inductive biases, relying instead on a causal assumption that the past causes the future. TDV functions by jointly training an image encoder and a motion encoder so that the current frame's representation plus the encoded motion equals the next frame's representation. Despite not leveraging any strong inductive biases, TDV matches state-of-the-art recipes on dense spatial tasks, laying the foundation for representation learning without strong assumptions.