Metraipolis › archive › dc_c4de17ea5e10

Evolving Memory, Learning to Meta-Control: Test-Time Adaptation for Experience-augmented LLM Agents — arXiv:2611.01747

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.01747
License (as archived)
CC0-1.0
Archived
2026-11-03T11:06:43.291Z
SHA-256 of the archived text
3f30910c2f1496a7fcb47583f41b69de9ec919dc11e5dd745a337550d0c7a214
Stable URL of this page
https://metraipolis.com/documento/dc_c4de17ea5e10

How to cite

Evolving Memory, Learning to Meta-Control: Test-Time Adaptation for Experience-augmented LLM Agents — arXiv:2611.01747 — arXiv (en), https://arxiv.org/abs/2611.01747 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c4de17ea5e10 (Metraipolis archive dc_c4de17ea5e10, sha256 3f30910c2f1496a7fcb47583f41b69de9ec919dc11e5dd745a337550d0c7a214)
BibTeX
@misc{dcc4de17ea5e10,
  title = {Evolving Memory, Learning to Meta-Control: Test-Time Adaptation for Experience-augmented LLM Agents — arXiv:2611.01747},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.01747},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c4de17ea5e10; sha256 3f30910c2f1496a7fcb47583f41b69de9ec919dc11e5dd745a337550d0c7a214},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01747. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Self-evolving agents accumulate experience from task streams, but existing frameworks typically rely on fixed pipelines to retrieve and utilize this experience, neglecting when and how to deploy it. We introduce Online Meta-Control of Experience (OMCE), a framework that updates experience memory and learns a meta-control policy from task feedback during test-time adaptation. OMCE employs a lightweight contextual-bandit meta-controller that uses compact retrieval signals to select among four cognitive actions: DIRECT, EPISODE, GUIDE, and DISTILL. This routing mechanism operationalizes three coupled decisions: whether to invoke experience, which memory format to utilize, and when to consolidate new guidance. OMCE evolves online, leveraging task feedback to update the memory bank and refine the routing policy via contextual bandit updates. Experiments on four online adaptation benchmarks across four backbones show that OMCE improves first-attempt success by up to 20.5 percentage points over the strongest compared baseline, while reducing per-task token use by a factor of 5.7 to 62.5 relative to ACE with GPT-5.2. Controlled ablations further support the benefits of learned routing and continued routing policy adaptation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2611.01747 — 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 2026-11-03T10:59:46.046Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).