Agents' Room: Narrative Generation through Multi-step Collaboration
Source: https://arxiv.org/abs/2410.02603
Search results attributed the work to Google DeepMind; that was not confirmed on the arXiv page and is not recorded as fact.
Structured fields
{
"archived": false,
"revision": {
"ref": "v2 (submitted 14 Mar 2025); comments: Published as a conference paper at ICLR 2025",
"pinned": true,
"source_url": "https://arxiv.org/abs/2410.02603"
},
"checked_at": "2026-10-17T00:00:00.000Z",
"usefulness": "ICLR 2025 paper decomposing story writing into subtasks handled by specialized agents (inspired by narrative theory) and introducing the Tell Me A Story prompt/story dataset and a long-narrative evaluation framework. Relevant to story-development pipelines; only the abstract was read and the dataset link was not found.",
"source_type": "paper",
"availability": "available",
"source_owner": "Fantine Huot, Reinald Kim Amplayo, Jennimaria Palomaki, Alice Shoshana Jakobovits, Elizabeth Clark, Mirella Lapata",
"next_check_at": "2026-10-31T00:00:00.000Z",
"known_limitations": [
"No code or dataset link on the abs page; availability and license of Tell Me A Story are unknown.",
"Only the abstract page was read."
],
"upstream_updated_at": "2025-03-14T17:09:03.000Z"
}Evidence claims
- [upstream_claim] Abstract: the framework decomposes narrative writing into subtasks tackled by specialized agents. · locator: Abstract source
- [upstream_claim] Abstract: introduces Tell Me A Story, a dataset of complex writing prompts and human-written stories. · locator: Abstract source
- [sourced_fact] Comments field: published at ICLR 2025. · locator: Comments field source
- [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| documentation | CC-BY-4.0 | permitted | permitted | permitted | unknown | arXiv abs page shows a 'view license' link to Creative Commons Attribution 4.0 for the paper text; attribution required. Says nothing about datasets or code. |