titleHoneycomb: Constant-Size Scene Memory Representation for Video World Models
authorsJack Wei Lun Shi, Kaichen Zhou, Haoyu Chen, Yufeng Weng, Keane Ong, Ruojin Cai, Hang Hua, Justin K. W. Yeoh, Mengyu Wang
date2026-09-30
venuearXiv (preprint, not peer reviewed)
idarXiv:2609.37690
upvotes3 (Hugging Face Daily Papers, 2026-11-01)
linksabs pdf hf code
primerWorld models
abstractVideo world models require persistent scene memory to maintain consistency during long-horizon video generation. Existing spatial memories accumulate RGB observations or latent features, increasing storage requirements as generation proceeds. We introduce Honeycomb, a video world model built on HexMemory, our proposed low-rank representation for storing scene features in a fixed-size memory with a total of six spatial and spatiotemporal planes. A feed-forward writer maps each generated chunk into new plane features. As the spatial coverage or temporal range expands, we warp the previous planes while preserving their dimensions, then fuse them with the new features through confidence-weighted pooling and a learned residual correction. A reader retrieves latents from HexMemory to condition subsequent video generation. The writer processes only observations from the new chunk, avoiding per-scene optimization and repeated processing of the full history. Experiments on WorldScore and RealEstate10K demonstrate strong video generation quality and robust revisit consistency while keeping HexMemory feature storage constant throughout generation. Code and additional visualizations are available on our project page at https://jackswl.github.io/honeycomb/.

cite

@misc{shi2026honeycomb,
  title={Honeycomb: Constant-Size Scene Memory Representation for Video World Models},
  author={Jack Wei Lun Shi and Kaichen Zhou and Haoyu Chen and Yufeng Weng and Keane Ong and Ruojin Cai and Hang Hua and Justin K. W. Yeoh and Mengyu Wang},
  year={2026},
  eprint={2609.37690},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2609.37690}
}