Metraipolis › archive › dc_3ee2ed2d5f29

ATC: Agent Trajectory Control for Role-Contract Enforcement in Black-Box Multi-Agent Systems — arXiv:2611.04321

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04321
License (as archived)
CC0-1.0
Archived
2026-11-04T18:21:38.100Z
SHA-256 of the archived text
32817fb5f4c995df97c006e8bce3eda9f86c5452623793ad3a346ed6fccc84c7
Stable URL of this page
https://metraipolis.com/documento/dc_3ee2ed2d5f29

How to cite

ATC: Agent Trajectory Control for Role-Contract Enforcement in Black-Box Multi-Agent Systems — arXiv:2611.04321 — arXiv (en), https://arxiv.org/abs/2611.04321 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3ee2ed2d5f29 (Metraipolis archive dc_3ee2ed2d5f29, sha256 32817fb5f4c995df97c006e8bce3eda9f86c5452623793ad3a346ed6fccc84c7)
BibTeX
@misc{dc3ee2ed2d5f29,
  title = {ATC: Agent Trajectory Control for Role-Contract Enforcement in Black-Box Multi-Agent Systems — arXiv:2611.04321},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04321},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3ee2ed2d5f29; sha256 32817fb5f4c995df97c006e8bce3eda9f86c5452623793ad3a346ed6fccc84c7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04321. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-agent LLM systems assign different responsibilities to different agents: one plans, another retrieves evidence, another verifies, another executes. This division of labor is only meaningful if each agent stays within the functions it is authorized to perform. Over many turns, however, an agent can produce a locally reasonable message that exercises authority belonging to another role: a verifier that begins planning, a retriever that asserts unsupported conclusions, an executor that changes state no one approved. We formalize this as role drift, a turn-level breach of an agent's role contract. We introduce Agent Trajectory Control (ATC), which makes each role's authority explicit as a contract and enforces it at the point where a message would enter the shared conversation. ATC scores a proposed output for out-of-role deviation, leakage into the functions of other roles, impermissible speech acts, and logical contract breaches, then computes a leakage-regularized target that constrains regeneration; the revised message is re-checked before admission. Enforcement acts on the message rather than the generator, so ATC governs agents whose internal representations are unavailable. Across homogeneous and heterogeneous teams, ATC improves task performance and role consistency while lowering role-drift energy relative to strong baselines spanning prompting, memory, workflow control, and activation editing, with the largest gains under explicit role-confusion stress tests and in larger, more diverse teams.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2611.04321 — 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 2026-11-04T18:20:56.485Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).