Metraipolis › archive › dc_29ca2dca7772

The World Won't Wait: Evaluating and Teaching Agents Under a Live Clock — arXiv:2610.20665

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.20665
License (as archived): CC0-1.0
Archived: 2026-10-16T08:55:12.838Z
SHA-256 of the archived text: e3b68d48d23e602659c1c3f23e0b1b2793dc63f093cc4339ffe1e8dbb046d80f
Stable URL of this page: https://metraipolis.com/documento/dc_29ca2dca7772

How to cite

The World Won't Wait: Evaluating and Teaching Agents Under a Live Clock — arXiv:2610.20665 — arXiv (en), https://arxiv.org/abs/2610.20665 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_29ca2dca7772 (Metraipolis archive dc_29ca2dca7772, sha256 e3b68d48d23e602659c1c3f23e0b1b2793dc63f093cc4339ffe1e8dbb046d80f)

BibTeX

@misc{dc29ca2dca7772,
  title = {The World Won't Wait: Evaluating and Teaching Agents Under a Live Clock — arXiv:2610.20665},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20665},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_29ca2dca7772; sha256 e3b68d48d23e602659c1c3f23e0b1b2793dc63f093cc4339ffe1e8dbb046d80f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20665. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language model (LLM) agents can solve complex interactive tasks, yet reasoning and tool calls take time. In continuously changing environments, an otherwise correct action may arrive too late. We introduce RealtimeBench, with 180 tasks across six game types, to evaluate agents while games continue running during inference. Paired evaluations show that GPT-6.1-sol outperforms the faster Jev when games are paused, but falls behind when they keep running. We propose Two-Timescale Offline Recursive Self-Improvement (TORSI): a slow teacher studies failures and tests corrections in paused games between attempts, then writes reusable instructions for a fast student. The student acts without teacher calls during gameplay, and validation scores determine which updates are adopted. Two rounds raise Jev's average test score from 13.51 to 24.19 on a 0--100 progress scale, improving all six game types. The instructions also improve average scores for three other models without retuning. These findings point toward real-time intelligence: enabling powerful LLM agents to remain effective in a continuously changing world by organizing deliberation, learning, and action across timescales.

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/2610.20665 — 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 2026-10-16T08:44:17.636Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).