leggedrobotics/navitrace
NaviTrace ๐ Project ๐ Paper ๐ป Code ๐ Leaderboard NaviTrace is a novel VQA benchmark for VLMs that evaluates models on their embodiment-specific understanding of navigation across challenging real-world scenarios. Key Features โ๏ธ Core Task: Given a real-world image in first-person perspective, a language instruction, and an embodiment type, models shouldโฆ See the full description on the dataset page: https://huggingface.co/datasets/leggedrobotics/navitrace.
| Author / organization | leggedrobotics |
|---|---|
| License (as declared) | cc-by-4.0 |
| Downloads | 1.2K (as reported by Hugging Face) |
| Likes | 7 |
| Created | 3 October 2025 |
| Last updated at source | 28 November 2025 |
Popularity, last 3 days
1,222 now (-191 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 17 Oct 2026 Description: NaviTrace ๐ Project ๐ Paper ๐ป Code ๐ Leaderboard NaviTrace is a novel VQA benchmark for VLMs that evaluates models on their embodiment-specific understanding of navigation across challenging real-world scenarios. Key Features โ๏ธ Core Ta โ NaviTrace ๐ Project ๐ Paper ๐ป Code ๐ Leaderboard NaviTrace is a novel VQA benchmark for VLMs that evaluates models on their embodiment-specific understanding of navigation across challenging real-world scenarios. Key Features โ๏ธ Core Ta
Changes detected by Silytics between refreshes of the source.
More from leggedrobotics
- leggedrobotics/grand_tour_datasetDataset
- leggedrobotics/hoi-retargetDataset
- leggedrobotics/egohtrDataset
- leggedrobotics/funthor-datasetDataset
- leggedrobotics/wildosDataset
- leggedrobotics/ViserDexSplatsDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.