# SAGE: Work-zone path prediction under temporary traffic control combining vision-language and geometric frameworks

- Source: arXiv cs.RO
- Published: 2026-10-26T04:00:00.000Z
- RoboHot: https://robohot.news/en/items/vpbs97rd6i230u5zl0njptqop
- Original: https://arxiv.org/abs/2610.31835

## Summary

The authors propose the SAGE (Semantic-to-Action Grounded Execution) framework, which uses trajectories output by a vision-language model (VLM) as preferences over complete geometric candidate paths, and combines each candidate's geometric score with its agreement with the VLM trajectory through rank fusion.