Metraipolis › archive › dc_2f27d29ec2f9
WorldBench: Evaluating LLMs on Three.js Voxel World Generation — arXiv:2610.10622
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.10622
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T05:53:33.868Z
- SHA-256 of the archived text
6f2ea8e42b6dc9eae4b1fe9bb32ded9512429c24f908b8ab12b0c1db39f7032d- Stable URL of this page
https://metraipolis.com/documento/dc_2f27d29ec2f9
How to cite
WorldBench: Evaluating LLMs on Three.js Voxel World Generation — arXiv:2610.10622 — arXiv (en), https://arxiv.org/abs/2610.10622 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2f27d29ec2f9 (Metraipolis archive dc_2f27d29ec2f9, sha256 6f2ea8e42b6dc9eae4b1fe9bb32ded9512429c24f908b8ab12b0c1db39f7032d)
BibTeX
@misc{dc2f27d29ec2f9,
title = {WorldBench: Evaluating LLMs on Three.js Voxel World Generation — arXiv:2610.10622},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.10622},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2f27d29ec2f9; sha256 6f2ea8e42b6dc9eae4b1fe9bb32ded9512429c24f908b8ab12b0c1db39f7032d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10622. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models can now write complete, interactive 3D worlds as code, but grading those worlds automatically is unreliable. Existing judges take one view of the output: a vision-language model scores a few rendered snapshots, or a language model reads the source. On worlds written by five frontier models we find that the two views disagree on 32% of required items, mostly code that no frame shows, and that fixed views miss small close-up contents. We present WorldBench, a benchmark and judge for open-ended, LLM-generated this http URL worlds. From one prompt describing a floating voxel island with ten biomes, physics, and day/night and seasonal cycles, the judge explores the running world, controlling its clock, orbiting it, and sending a navigator agent to frame each biome, and reads the code for what it sees. Neither channel is trusted on its own: a code quote counts only if it is text the source contains, and visual claims are checked against measured pixels where the property is measurable. A mutation test, in which we remove features by construction, shows that code-only judging gives full credit to four of five removed features, because their code remains in the file. Our judge cuts the points kept on removed features by a third (5.44 to 3.55 of 7.11), and what it still credits is mostly code that exists but never runs. We evaluate five frontier models: Claude Fable 5.1, GPT-6 Astra, Kimi K3, Grok 4.7 and Gemini 3.1 Pro. Code, prompt, tests and judge configuration are available at this https URL Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Graphics (cs.GR) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2610.10622 — 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 2026-10-09T05:47:01.214Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).