Metraipolis › archive › dc_cff7f80274b7

Decision Context: Persistent Behavioral Learning for AI Agents — arXiv:2611.08644

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.08644
License (as archived)
CC0-1.0
Archived
2026-11-09T22:33:38.498Z
SHA-256 of the archived text
67cca0b31559354f6492f02601f6289e830506d0866d751725bf4201fc77994b
Stable URL of this page
https://metraipolis.com/documento/dc_cff7f80274b7

How to cite

Decision Context: Persistent Behavioral Learning for AI Agents — arXiv:2611.08644 — arXiv (en), https://arxiv.org/abs/2611.08644 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cff7f80274b7 (Metraipolis archive dc_cff7f80274b7, sha256 67cca0b31559354f6492f02601f6289e830506d0866d751725bf4201fc77994b)
BibTeX
@misc{dccff7f80274b7,
  title = {Decision Context: Persistent Behavioral Learning for AI Agents — arXiv:2611.08644},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.08644},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cff7f80274b7; sha256 67cca0b31559354f6492f02601f6289e830506d0866d751725bf4201fc77994b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.08644. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM agents do not learn from corrections across sessions. A user fixes the same mistake at the start of every conversation. Existing agent memory systems primarily handle facts, such as preferences, entities, and conversational history, rather than procedural knowledge. We present Forge Decision Context (forge_dc), a system that captures behavioral signals from human corrections and self-assessed tool-call failures, compiles them into imperative RULE directives, and injects them at the decision points where they apply. We evaluate forge_dc alongside four predecessor approaches on DC-Bench, a new benchmark of 71 scenarios across nine behavioral dimensions, built from real executed agent traces in a deterministic mocked-tool environment. forge_dc reaches 95% compliance, up from 55% with naive retrieval, while injecting approximately 10 times fewer tokens per query. Across the five-stage progression, the strongest systems represent behavioral memory as explicit directives and deliver it at the decision points where it applies. The results identify representation, activation, and placement as central design choices for persistent behavioral learning.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.08644 — 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 2026-11-09T22:30:47.249Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).