Metraipolis › archive › dc_d5f00cab56bc

When Successful Memories Mislead Embodied Agents:Memory Adaption For Task-Conditioned Execution — arXiv:2609.35808

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.35808
License (as archived)
CC0-1.0
Archived
2026-10-07T16:24:41.286Z
SHA-256 of the archived text
86ab77a55fe7b3e274eb8e77caf0831432ed604bcb03269a516662fad9fe462c
Stable URL of this page
https://metraipolis.com/documento/dc_d5f00cab56bc

How to cite

When Successful Memories Mislead Embodied Agents:Memory Adaption For Task-Conditioned Execution — arXiv:2609.35808 — arXiv (en), https://arxiv.org/abs/2609.35808 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d5f00cab56bc (Metraipolis archive dc_d5f00cab56bc, sha256 86ab77a55fe7b3e274eb8e77caf0831432ed604bcb03269a516662fad9fe462c)
BibTeX
@misc{dcd5f00cab56bc,
  title = {When Successful Memories Mislead Embodied Agents:Memory Adaption For Task-Conditioned Execution — arXiv:2609.35808},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35808},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d5f00cab56bc; sha256 86ab77a55fe7b3e274eb8e77caf0831432ed604bcb03269a516662fad9fe462c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35808. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Experience reuse can reduce repeated exploration in embodied agents, but a trajectory that succeeded previously may be unsuitable for the current execution context. Existing memory systems pri marily optimize construction and retrieval; semantic relevance and historical success therefore remain insufficient when retrieved ex perience contains incompatible actions or an inappropriate level of structure. We introduce Memory Adaptation for Task-Conditioned Execution (MATE), a deterministic post-retrieval procedure that converts trajectories into execution-oriented memory. MATE re moves obsolete control context, extracts condition-action-effect transitions, applies verified action normalization, selects a task dependent representation, and serializes the result under a fixed budget without additional LLM inference. On 134 ALFWorld tasks, MATE achieves task success rates of 81.3% and 93.3% with Qwen2.5-14B and 72B while using approximately one-tenth of the tokens required by raw trajectories. Controlled comparisons show that verified action normalization is the principal mechanism by which MATE restores the utility of retrieved experience, support ing memory adaptation as a distinct stage between retrieval and embodied execution.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.35808 — 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 2026-10-07T15:54:53.708Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).