Metraipolis › archive › dc_5bc79545ff10
Mem4Act: Learning Effective Latent Memory for Long-Horizon Robotic Manipulation — arXiv:2610.22323
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.22323
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T10:40:11.112Z
- SHA-256 of the archived text
7a1f888b217dc82fed3002ba1f74c7d5ab1f238657c8212c44c8556efcb219e0- Stable URL of this page
https://metraipolis.com/documento/dc_5bc79545ff10
How to cite
Mem4Act: Learning Effective Latent Memory for Long-Horizon Robotic Manipulation — arXiv:2610.22323 — arXiv (en), https://arxiv.org/abs/2610.22323 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5bc79545ff10 (Metraipolis archive dc_5bc79545ff10, sha256 7a1f888b217dc82fed3002ba1f74c7d5ab1f238657c8212c44c8556efcb219e0)
BibTeX
@misc{dc5bc79545ff10,
title = {Mem4Act: Learning Effective Latent Memory for Long-Horizon Robotic Manipulation — arXiv:2610.22323},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.22323},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5bc79545ff10; sha256 7a1f888b217dc82fed3002ba1f74c7d5ab1f238657c8212c44c8556efcb219e0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22323. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-horizon robotic manipulation fundamentally requires memory: agents must retain task-relevant information from past observations and interactions and use it to guide future actions. Existing approaches typically preserve raw historical observations, compressed visual tokens, or language summaries, but these representations either expose the action policy to increasingly long histories or risk discarding task-relevant temporal, semantic, and perceptual information. We introduce Mem4Act, a framework for learning effective latent memory as a fixed-size representation explicitly optimized for long-horizon robotic control. Mem4Act addresses three coupled challenges: how to efficiently cover long histories, what information memory should retain, and how to ensure the policy actually relies on memory. To this end, it organizes history via multi-granularity temporal aggregation, preserves complementary semantic and perceptual information through question-answering supervision and key-frame DINO feature reconstruction, and employs memory-first two-stage training to encourage greater use of memory by the action policy. The resulting memory captures both high-level task state and fine-grained perceptual evidence while providing a bounded interface to the action policy. Across RoboMME, RMBench, and four real-world manipulation tasks, Mem4Act achieves state-of-the-art performance on memory-dependent control, with success rates of 49.88%, 61.0%, and 81.5%, respectively. Extensive ablations further demonstrate the importance of temporal aggregation, complementary memory supervision, and memory-first training. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2610.22323 — 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 2026-10-16T08:52:17.707Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).