Metraipolis › archive › dc_09bb5af7c312
GaussMemory: Task-Driven 3D Gaussian Scene Memory for Long-Horizon Robotic Manipulation — arXiv:2608.14986
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.14986
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T17:55:46.782Z
- SHA-256 of the archived text
5ba94001f8d039b39bbb5c1f05810811280322826e8ff7c95b7f319e5f009ebc- Stable URL of this page
https://metraipolis.com/documento/dc_09bb5af7c312
How to cite
GaussMemory: Task-Driven 3D Gaussian Scene Memory for Long-Horizon Robotic Manipulation — arXiv:2608.14986 — arXiv (en), https://arxiv.org/abs/2608.14986 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_09bb5af7c312 (Metraipolis archive dc_09bb5af7c312, sha256 5ba94001f8d039b39bbb5c1f05810811280322826e8ff7c95b7f319e5f009ebc)
BibTeX
@misc{dc09bb5af7c312,
title = {GaussMemory: Task-Driven 3D Gaussian Scene Memory for Long-Horizon Robotic Manipulation — arXiv:2608.14986},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.14986},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_09bb5af7c312; sha256 5ba94001f8d039b39bbb5c1f05810811280322826e8ff7c95b7f319e5f009ebc},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14986. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-horizon robotic manipulation fundamentally relies on persistent spatial memory. However, existing 3D memory systems function merely as passive recorders: they store observations using fixed, hand-crafted rules, treating every scene element--whether a critical grasp target or an irrelevant background wall--with equal importance. In this paper, we propose a paradigm shift from passive storage to active, task-driven spatial memory. We argue that a robot's memory should not simply record what it sees, but actively learn how to remember--discovering which objects to track precisely, how aggressively to update them, and what to discard, all learned end-to-end without hand-designed rules. Crucially, this active paradigm is realized by unifying memory update and readout as two sides of the same cognitive process, enabling bidirectional flow where task needs shape update strategies and vice versa. To instantiate this vision, we introduce GaussMemory, which leverages 3D Gaussian Splatting as a persistent geometric substrate. On LIBERO, GaussMemory outperforms MemoryVLA on Goal and Long-10; on VLABench, it surpasses $\pi_0$-FAST by +5.2% (Track 1) and +6.0% (Track 6). Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.AI). Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.14986 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-18T16:29:13.972Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).