Metraipolis › archive › dc_772a105cfee5

Do Agents Still Follow the Rules? Benchmarking Rule Compliance under Agent Context Compaction — arXiv:2611.03223

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.03223
License (as archived)
CC0-1.0
Archived
2026-11-03T18:20:24.101Z
SHA-256 of the archived text
c7a9f112f0441fda85bacfec2198e06cb1b5176fca8965ff7fa4a267580c04b2
Stable URL of this page
https://metraipolis.com/documento/dc_772a105cfee5

How to cite

Do Agents Still Follow the Rules? Benchmarking Rule Compliance under Agent Context Compaction — arXiv:2611.03223 — arXiv (en), https://arxiv.org/abs/2611.03223 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_772a105cfee5 (Metraipolis archive dc_772a105cfee5, sha256 c7a9f112f0441fda85bacfec2198e06cb1b5176fca8965ff7fa4a267580c04b2)
BibTeX
@misc{dc772a105cfee5,
  title = {Do Agents Still Follow the Rules? Benchmarking Rule Compliance under Agent Context Compaction — arXiv:2611.03223},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03223},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_772a105cfee5; sha256 c7a9f112f0441fda85bacfec2198e06cb1b5176fca8965ff7fa4a267580c04b2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03223. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM agents compact long histories to fit their context windows, but the compacted context must still carry the operating rules that govern later actions. Prior work shows that violations rise after compaction, but measures compliance outside a complete agent harness and does not identify where each violation comes from. We introduce ARCBench(Agent Rule Compliance Benchmark), a configurable benchmark built on the open-source pi agent harness. Its tool-calling agent loop lets us measure rule compliance from actions taken during execution, bringing the evaluation closer to deployed agent workflows. Using ARCBench, we examine where violations after compaction come from, and how different compaction configurations and adversarial settings change their sources. We find that compaction rarely deletes the rule: most violations after compaction occur while the rule is still in the compacted history, with the agent treating a later request as permission, ignoring the rule, and proceeding anyway. Where violations come from depends on the configuration: how the rule is delivered decides whether the agent would break it even without compaction, while the compaction strategy and budget decide whether the rule survives. The strongest compactor-targeted attack more than doubles violations by corrupting the rule in the summary and changing how the agents behave, whereas the strongest defense restores the rule in most summary but still violates. Finally, retaining a rule does not guarantee compliance: compacted runs that retain the rule violate more than twice as often as runs with the full history.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2611.03223 — 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 2026-11-03T18:09:53.746Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).