Metraipolis › archive › dc_1820162aee11
JEPAhead: From Inpainting to Causal Forecasting for Urban Driving — arXiv:2611.08657
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.08657
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-09T22:34:28.645Z
- SHA-256 of the archived text
b7233fe6bdcc042a4f2d3c8fb81ae1651595aa68fd5f179c8c5d4c3fb0ceabf2- Stable URL of this page
https://metraipolis.com/documento/dc_1820162aee11
How to cite
JEPAhead: From Inpainting to Causal Forecasting for Urban Driving — arXiv:2611.08657 — arXiv (en), https://arxiv.org/abs/2611.08657 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1820162aee11 (Metraipolis archive dc_1820162aee11, sha256 b7233fe6bdcc042a4f2d3c8fb81ae1651595aa68fd5f179c8c5d4c3fb0ceabf2)
BibTeX
@misc{dc1820162aee11,
title = {JEPAhead: From Inpainting to Causal Forecasting for Urban Driving — arXiv:2611.08657},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.08657},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1820162aee11; sha256 b7233fe6bdcc042a4f2d3c8fb81ae1651595aa68fd5f179c8c5d4c3fb0ceabf2},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.08657. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We fine-tune a pre-trained V-JEPA1 predictor from bidirectional latent inpainting to action-free causal forecasting while keeping the encoder frozen. Using contiguous past observations, horizon-aligned targets, and group-level causal attention, the model forecasts eight future token groups, corresponding to sixteen frames, in one forward pass. On CARLA Town12, it retrieves the correct future among 591 candidates in 69.0% of cases, compared with 55.3% for persistence, with greater gains at longer horizons and on tokens exhibiting greater change. It also outperforms a compact predictor trained from scratch with six times fewer parameters. Semantic analysis indicates driving-relevant information in predicted features: a fixed linear probe trained on real encoder tokens achieves 0.16 mIoU on patches undergoing semantic-class changes, versus 0.13 for persistence. Vehicle-conditioned retrieval improves over persistence by 15 percentage points at the longest horizon. Pixel-space evaluation shows modest L1 gains but little improvement under LPIPS, motivating direct evaluation through latent and semantic measures alongside an oracle reconstruction reference. These findings support further investigation of latent forecasting for driving perception, while downstream utility remains unverified. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.08657 — 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 2026-11-09T22:30:47.302Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).