Metraipolis › archive › dc_f028ae0d2ae8
Shared Worlds, Private Minds: Structured Memory for Long-Form Writing as World Creation — arXiv:2609.32401
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.32401
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:00:10.729Z
- SHA-256 of the archived text
f4ea4f01d677be204d93edbe32aabcbdaae0efa1250927a530dcf89fec5f1ed1- Stable URL of this page
https://metraipolis.com/documento/dc_f028ae0d2ae8
How to cite
Shared Worlds, Private Minds: Structured Memory for Long-Form Writing as World Creation — arXiv:2609.32401 — arXiv (en), https://arxiv.org/abs/2609.32401 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f028ae0d2ae8 (Metraipolis archive dc_f028ae0d2ae8, sha256 f4ea4f01d677be204d93edbe32aabcbdaae0efa1250927a530dcf89fec5f1ed1)
BibTeX
@misc{dcf028ae0d2ae8,
title = {Shared Worlds, Private Minds: Structured Memory for Long-Form Writing as World Creation — arXiv:2609.32401},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.32401},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f028ae0d2ae8; sha256 f4ea4f01d677be204d93edbe32aabcbdaae0efa1250927a530dcf89fec5f1ed1},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32401. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
LLM agents that write long-form fiction need an explicit memory of the evolving storyworld to keep new events consistent with established facts. Such memory must keep heterogeneous narrative information distinct, integrate story developments across granularities, and recover dependencies that a writing request leaves implicit. We present NarraWorld, a structured memory system for long-form writing that treats memory construction as world creation. From a shared evidence-grounded graph, NarraWorld derives four connected views: world facts, per-character beliefs, open developments, and hypothetical branches (possible-world continuations). Hierarchical aggregation with atomic closure consolidates events into scenes, plotlines, and plots, keeping each higher-level node traceable to its constituent source spans. For retrieval, planned reconstruction infers a query's dependencies from the current narrative situation and a preview of memory, then assembles the relevant records within a token budget. Across three writing benchmarks, NarraWorld achieves the strongest aggregate results. Its memory also transfers to situated role-playing and largely preserves recall on a general-purpose long-term memory benchmark, paving the way for agents that sustain coherent storyworlds across diverse narrative tasks. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.32401 — 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 2026-10-07T20:29:02.326Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).