SoftVTBench: A Deformation-Aware Visuo-Tactile Dataset and Benchmark for Deformable-Object Manipulation
- 类型:arxiv
- 标识:2608.18701
- 链接:https://arxiv.org/abs/2608.18701
- 主分类:evaluation
- 形态:benchmark
- 被引:0
- 被引来源:Semantic Scholar
- S2被引:0
- 影响力被引:0
- TLDR:Results show that making touch available does not by itself ensure effective multimodal fusion, and SoftVTBench provides a common visuo-tactile resource for studying not only whether a policy succeeds, but how it physically interacts with deformable objects and when touch improves that interaction.
- 待LLM分类:否
- 标题中文:SoftVTBench:面向可形变物体操作的形变感知视触觉数据集与基准
- TLDR中文:实验表明,仅提供触觉本身并不能确保有效的多模态融合,SoftVTBench 为研究策略不仅能否成功,还在于其如何与可形变物体物理交互,以及触觉在何时改善这种交互,提供了统一的视触觉资源
- 来源文件:
- /inbox/tom/_candidates/2026-08-20-agent-rag-longcontext-candidates.json
- [S2 enrich]