超越汇点解释:基于流的VLA策略中的径向注意力
Beyond the Sink Interpretation: Radial Attention in Flow-Based VLA Policies
Yuxi Zheng, Yueqian Lin, Yudong Liu, Zijia Tang, Junyu Pan, Yiheng Wang, Hai Helen Li, Yiran Chen
AI 速读 基于摘要
作者在单个π0.5检查点上检验了流式VLA策略中首尾动作令牌的高注意力是否构成“汇点”,发现它们不符合汇点特征,而前缀BOS令牌才符合,且相对偏移路由能保持动作令牌的分散性和闭环成功率。
cs.RO
查看原文摘要
Prior work calls the high attention on the first and last action tokens of a flow-based VLA policy a radial action sink, and uses these tokens to set the execution horizon. In language models, a sink is a token with high incoming attention, a small value, and little effect on the output. We test whether that profile holds, not the horizon rule, in one $π_{0.5}$ checkpoint on five RoboCasa tasks. The two endpoints receive 5.4% of action-key mass, 1.4 percentage points above their uniform share, and rotary positional encoding produces this excess: directly in shallow layers and through upstream action-token RoPE deeper in the network. The endpoints do not fit the sink profile. Their value norms stay near the action-token mean in all 18 layers. On 250 paired episodes, masking them or zeroing their values changes success by less than 4 points, and replacing their value directions with those of another observation lowers it by 3.2 to 7.2 points across three donors, whereas the same replacement on two non-endpoint tokens in the executed prefix lowers it by 28.4 points. The token that does fit the sink profile on mass and value norm is the prefix BOS; on this checkpoint, zeroing its value raises success in two paired runs, by +13.2 points (95% CI [+7.0, +19.4]) and +6.0 points ([+0.2, +11.8]). Inside the action chunk, a relative-offset routing profile estimated on other tasks keeps 75% of action-token spread and keeps closed-loop success on three held-out tasks (42.7% against a 38.7% baseline), while uniform or shuffled routing drops success to 3-4%. High attention mass alone does not tell what a token does.