Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.10432 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914615282106368 |
|---|---|
| author | Hu, Zhaofeng Zhou, Sifan Zhang, Qinbo Xu, Rongtao Su, Qi Mendez-Mendz, Jorge Liang, Ci-Jyun |
| author_facet | Hu, Zhaofeng Zhou, Sifan Zhang, Qinbo Xu, Rongtao Su, Qi Mendez-Mendz, Jorge Liang, Ci-Jyun |
| contents | Vision-Language-Action (VLA) policies have emerged as a versatile paradigm for generalist robotic manipulation. However, precise object placement under compositional language remains challenging for end-to-end VLA policies. Slot-level placement requires reliable slot grounding and centimeter-level geometric precision. To this end, we propose AnySlot, a framework that reduces compositional complexity by introducing an explicit spatial visual goal between language grounding and control. AnySlot converts language into a visual goal by rendering a spatial marker at the intended slot, then executes this goal with a goal-conditioned VLA policy. This hierarchical design decouples high-level slot selection from low-level execution, improving semantic accuracy and spatial robustness. Furthermore, recognizing the lack of benchmarks for such precision-demanding tasks, we introduce SlotBench, a structured simulation benchmark with nine task categories for evaluating spatial reasoning in slot-level placement. Extensive experiments show that AnySlot significantly outperforms flat VLA baselines and modular grounding methods in zero-shot slot-level placement. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_10432 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AnySlot: Goal-Conditioned Vision-Language-Action Policies for Zero-Shot Slot-Level Placement Hu, Zhaofeng Zhou, Sifan Zhang, Qinbo Xu, Rongtao Su, Qi Mendez-Mendz, Jorge Liang, Ci-Jyun Robotics Vision-Language-Action (VLA) policies have emerged as a versatile paradigm for generalist robotic manipulation. However, precise object placement under compositional language remains challenging for end-to-end VLA policies. Slot-level placement requires reliable slot grounding and centimeter-level geometric precision. To this end, we propose AnySlot, a framework that reduces compositional complexity by introducing an explicit spatial visual goal between language grounding and control. AnySlot converts language into a visual goal by rendering a spatial marker at the intended slot, then executes this goal with a goal-conditioned VLA policy. This hierarchical design decouples high-level slot selection from low-level execution, improving semantic accuracy and spatial robustness. Furthermore, recognizing the lack of benchmarks for such precision-demanding tasks, we introduce SlotBench, a structured simulation benchmark with nine task categories for evaluating spatial reasoning in slot-level placement. Extensive experiments show that AnySlot significantly outperforms flat VLA baselines and modular grounding methods in zero-shot slot-level placement. |
| title | AnySlot: Goal-Conditioned Vision-Language-Action Policies for Zero-Shot Slot-Level Placement |
| topic | Robotics |
| url | https://arxiv.org/abs/2604.10432 |