Metraipolis › archive › dc_92bda26da7cd

The Horizon Gap: Planning, Memory, Execution, Training, and Evaluation for Long-Horizon LLM Agents — arXiv:2608.06663

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.06663
License (as archived): CC0-1.0
Archived: 2026-10-19T09:46:47.207Z
SHA-256 of the archived text: fee62124fc00845a963a8e17dfb434d49ff8490e73e9b1913de1ce1006eb5068
Stable URL of this page: https://metraipolis.com/documento/dc_92bda26da7cd

How to cite

The Horizon Gap: Planning, Memory, Execution, Training, and Evaluation for Long-Horizon LLM Agents — arXiv:2608.06663 — arXiv (en), https://arxiv.org/abs/2608.06663 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_92bda26da7cd (Metraipolis archive dc_92bda26da7cd, sha256 fee62124fc00845a963a8e17dfb434d49ff8490e73e9b1913de1ce1006eb5068)

BibTeX

@misc{dc92bda26da7cd,
  title = {The Horizon Gap: Planning, Memory, Execution, Training, and Evaluation for Long-Horizon LLM Agents — arXiv:2608.06663},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.06663},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_92bda26da7cd; sha256 fee62124fc00845a963a8e17dfb434d49ff8490e73e9b1913de1ce1006eb5068},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.06663. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Frontier language models solve reasoning problems in a single forward pass that would have been research contributions years ago, yet fail at multi-hour tasks: losing track of earlier decisions, declaring half-finished work done, or drifting from goals. We call this the horizon gap and survey 1,547 arXiv papers (2024-2026) collected via systematic seed harvest with a disclosed 26.8% bleed filter, extended by targeted supplementation. We disambiguate three routinely conflated properties: long-horizon (task property: required steps), long-context (model property: token capacity), and long-term memory (system property: persistence across steps/sessions). We organize the corpus into six categories tracking a long-horizon task's lifecycle -- planning, memory, execution, training, evaluation, and foundations/safety -- crossed with an axis capturing where horizons are carried (within-context, within-task-beyond-context, or cross-task-persistent). Across all categories, we find the same pattern: outcome-only signals grow uninformative as horizons lengthen, and the field's response -- whether process reward models, credit assignment, or trajectory-level diagnostics -- manufactures denser step-level signals. We treat critical and diagnostic literature as first-class threads throughout, arguing that segregating critique from method would routinely split single papers across chapters. We close by naming open measurement problems: decomposing model versus harness capability, managing correlated bias in process-level signals used for both training and evaluation, and whether long-horizon reliability admits general predictive theory.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-10 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.06663 — 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 2026-10-19T09:46:37.211Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).