Metraipolis › archive › dc_fddda2a0d6b3

Argus: A General-Purpose Agentic Reasoning Runtime for Long-Horizon Tasks — arXiv:2608.05144

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.05144
License (as archived)
CC0-1.0
Archived
2026-10-31T11:13:19.259Z
SHA-256 of the archived text
d091437c804f818a0dd88026e40d30b57e5e8a78475236f57f4b4e8d776c8983
Stable URL of this page
https://metraipolis.com/documento/dc_fddda2a0d6b3

How to cite

Argus: A General-Purpose Agentic Reasoning Runtime for Long-Horizon Tasks — arXiv:2608.05144 — arXiv (en), https://arxiv.org/abs/2608.05144 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fddda2a0d6b3 (Metraipolis archive dc_fddda2a0d6b3, sha256 d091437c804f818a0dd88026e40d30b57e5e8a78475236f57f4b4e8d776c8983)
BibTeX
@misc{dcfddda2a0d6b3,
  title = {Argus: A General-Purpose Agentic Reasoning Runtime for Long-Horizon Tasks — arXiv:2608.05144},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.05144},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fddda2a0d6b3; sha256 d091437c804f818a0dd88026e40d30b57e5e8a78475236f57f4b4e8d776c8983},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05144. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Long-horizon reasoning requires an agentic runtime that can persist when evidence supports its current approach and pivot when measurements reveal failure, hidden constraints, or a misspecified objective. We present Argus, a persistent, self-evolving runtime in which Manager, Planner, Engineer, and Reviewer execute bounded missions over durable project state. Argus separates stable user intent from operational objectives, constraints, and verification criteria, and admits memories, skills, procedures, verifiers, routing decisions, and rejected routes only after role-owned review and, when available, task-native verification. Model weights remain fixed; self-evolution occurs through persistent runtime state and control policy, with autonomous execution between operator-owned escalation points. Across seven GPT-5.5 benchmark arenas, Argus achieves about 78% on SWE-Bench Pro versus 59% for Direct Copilot while using 1.41 times the aggregate tokens. After verification-gated self-evolution, mature SWE-Bench waves use 21% fewer solve-input tokens and 15% less active workflow time per task than startup waves, while recording 34 verifier recoveries and 22 strict review-loop rescues. Argus also reaches 76.8% on AARRI-Bench and a 28.0-point gap on mathematical data synthesis, with competitive GPU-kernel and language-model-training results. Beyond benchmarks, an optimized RWKV6 kernel was merged upstream; a multi-day mathematics campaign retained falsified routes and proof-backed frontier updates; and six paper pipelines completed 254 missions with 16 stage rollbacks. These results show that a fixed-weight, self-evolving harness can revise, recover, and accumulate verified approaches while producing structured trajectories for future supervised and reinforcement learning.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-06 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.05144 — 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 2026-10-31T11:02:13.444Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).