AI Paper Atlas

Research map / Video / Video generation

MicroVerse: A Preliminary Exploration Toward a Micro-World Simulation

Rongsheng Wang, Minghao Wu, Hongru Zhou, Zhihan Yu, Zhenyang Cai, Junying Chen, Benyou Wang

ICLR 2026 · Video › Video generation · approach: World-model video generation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fine-tune on MicroSim-10KWan2.1-T2V-1.3B · MicroWorldBenchscientific fidelity: 40.3 -> 43.0Improved
fine-tune on MicroSim-10KWan2.1-T2V-1.3B · MicroWorldBenchvisual quality: 71.8 -> 68.5Hurt
fine-tune on MicroSim-10KWan2.1-T2V-1.3B · MicroWorldBenchinstruction following: 50.1 -> 49.3Hurt
fine-tune on MicroSim-10KWan2.1-T2V-1.3B · MicroWorldBench subcellular-levelrubric score: 52.4 -> 53.3Improved
add general-domain data to MicroSim-10K (mixed-domain training) (ablation)MicroVerse 1.3B · MicroWorldBenchscientific fidelity: 43.0 -> 44.1Improved
add general-domain data to MicroSim-10K (mixed-domain training) (ablation)MicroVerse 1.3B · MicroWorldBenchvisual quality: 68.5 -> 74.4Improved
add general-domain data to MicroSim-10K (mixed-domain training) (ablation)MicroVerse 14B · MicroWorldBenchscientific fidelity: 45.4 -> 48.3Improved
fine-tune 14B only on MicroSim-10KWan2.1-14B · MicroWorldBenchvisual quality: 86.0 -> 82.7Hurt
fine-tune 14B only on MicroSim-10KWan2.1-14B · MicroWorldBenchscientific fidelity: 42.7 -> 45.4Improved
scale model from 1.3B to 14B (baseline Wan2.1)Wan2.1 · MicroWorldBenchscientific fidelity: 40.3 -> 42.7No clear effect
scale model from 1.3B to 14B (baseline Wan2.1)Wan2.1 · MicroWorldBenchvisual quality: 71.8 -> 86.0Improved
use expert-filtered MicroSim-10K instead of raw uncleaned data (ablation)MicroVerse 1.3B · MicroWorldBenchinstruction following: 44.5 -> 49.3Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy