Add UrbanGround dataset card and license

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 UrbanGround Contributors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -0,0 +1,197 @@
+---
+pretty_name: UrbanGround Tasks
+license: mit
+language:
+- en
+task_categories:
+- visual-question-answering
+- reinforcement-learning
+- robotics
+tags:
+- embodied-ai
+- multimodal
+- spatial-reasoning
+- navigation
+- agent-evaluation
+- 3d-city
+- hong-kong
+- arxiv:2608.27456
+size_categories:
+- n<1K
+configs:
+- config_name: default
+  data_files:
+  - split: test
+    path: data/benchmark.jsonl
+  - split: auxiliary
+    path: data/auxiliary_variants.jsonl
+  default: true
+---
+
+# UrbanGround Tasks
+
+UrbanGround Tasks is the evaluation dataset for [UrbanGround: From Local Perception to Spatial Agency in a Real-Scale City](https://huggingface.co/papers/2608.27456). It contains georegistered task definitions for evaluating multimodal agents in the UrbanGround real-scale 3D replica of Hong Kong.
+
+- Paper: [arXiv:2608.27456](https://arxiv.org/abs/2608.27456)
+- Code and desktop applications: [UrbanGround/UrbanGround](https://github.com/UrbanGround/UrbanGround)
+- Interactive project page: [urbanground.github.io](https://urbanground.github.io/)
+
+## Quick start
+
+```python
+from datasets import load_dataset
+
+benchmark = load_dataset(
+    "jometeorie/urbanground-tasks",
+    split="test",
+)  # 810 paper benchmark instances
+
+print(benchmark[0])
+```
+
+To also inspect the 28 non-benchmark legacy variants, load the complete
+`DatasetDict` and use its `auxiliary` split:
+
+```python
+dataset = load_dataset("jometeorie/urbanground-tasks")
+auxiliary = dataset["auxiliary"]
+```
+
+The `test` split is the complete evaluation benchmark. It is not intended as a
+training/test partition. Reference answers and evaluator-only fields are
+intentionally public for reproducible evaluation; leaderboard runs must control
+which fields are exposed to the evaluated agent.
+
+## Why 700 stored tasks become 810 benchmark instances
+
+The released application stores **700 unique benchmark source definitions** as JSON files. Level 5 changes the execution conditions of already validated navigation tasks instead of duplicating their geometry:
+
+- 30 Constrained Navigation instances are reused for Dynamic Road-Closure Replanning.
+- 80 Long-Range Goal Navigation instances are reused for Navigation among Pedestrians.
+
+This produces the **810 benchmark instances reported in the paper**. The `test` split materializes those 110 Level 5 views as self-contained rows and records their origin in `source_task_id`, `is_derived`, and `derivation`.
+
+The application package also contains 28 legacy environment and diagnostic
+variants. They retain their original `variant_prefix` and `source_task_id`, but
+the current application does not define a stable public condition schema for
+those prefixes. They are therefore kept in the separate `auxiliary` split and
+excluded from all benchmark totals.
+
+## Benchmark composition
+
+| Level | Capability | Subtask | Abbr. | Instances |
+|---:|---|---|:---:|---:|
+| 1 | Local Environment Understanding | Visual Recognition | VR | 80 |
+| 1 | Local Environment Understanding | Orientation Understanding | OU | 60 |
+| 1 | Local Environment Understanding | Active Exploration Questions | AEQ | 80 |
+| 2 | Navigation under Explicit Instructions | Short-Range Goal Navigation | SGN | 80 |
+| 2 | Navigation under Explicit Instructions | Long-Range Goal Navigation | LGN | 80 |
+| 2 | Navigation under Explicit Instructions | Instructional Navigation | IN | 50 |
+| 2 | Navigation under Explicit Instructions | Constrained Navigation | CN | 30 |
+| 3 | Exploration under Implicit Instructions | Place-Type Search | PTS | 60 |
+| 3 | Exploration under Implicit Instructions | Implicit Intent Inference | III | 60 |
+| 4 | Multi-Task Planning | Time-Window Scheduling | TWS | 60 |
+| 4 | Multi-Task Planning | Multi-Stop Route Planning | MSP | 60 |
+| 5 | Dynamic Environment Interaction | Dynamic Road-Closure Replanning | DCR | 30 |
+| 5 | Dynamic Environment Interaction | Navigation among Pedestrians | NP | 80 |
+|  |  | **Total** |  | **810** |
+
+## Data organization
+
+```text
+data/
+├── benchmark.jsonl             # 810 rows used by the paper
+└── auxiliary_variants.jsonl    # 28 non-benchmark variants
+metadata/
+└── statistics.json             # machine-readable counts
+raw/
+├── urbanground-app-task-files-v1.0.0.zip
+├── task-files.sha256
+└── manifest.json
+scripts/
+└── build_dataset.py
+```
+
+The raw ZIP preserves the application's original one-file-per-task representation (`{"task": {...}}`) for direct placement beside a desktop build. The JSONL files flatten that wrapper, use `snake_case` field names, and add human-readable benchmark metadata for the Dataset Viewer.
+
+## Important fields
+
+| Field | Description |
+|---|---|
+| `id` | Stable task instance identifier. |
+| `task_type_id` | Integer task type used by the UrbanGround application. |
+| `task_type`, `task_abbreviation` | Human-readable subtask name and paper abbreviation. |
+| `capability_level`, `capability_name` | Level 1–5 and its capability group. |
+| `release_role` | `stored_base`, `derived_level_5`, or `auxiliary_variant`. |
+| `is_derived`, `derivation` | Materialized Level-5 provenance. |
+| `source_task_id`, `variant_prefix` | Referenced stored task and retained legacy variant code, when applicable. |
+| `start_point`, `end_point`, `qa_start_point`, etc. | WGS84 positions; `height` is in meters and `yaw` is in degrees. |
+| `qa_options`, `qa_answer_index`, `qa_answer_text` | Multiple-choice candidates and the reference answer. |
+| `restricted_zones` | Closed-road geometry for constrained and dynamic replanning. |
+| `schedule`, `multi_route_targets` | Ordered or unordered multi-destination task definitions. |
+| `notes` | Annotator-only notes; these must not be sent to the evaluated agent. |
+
+Coordinates and reference fields are evaluator state, not a model-visible observation. In particular, `qa_answer_index`, `qa_answer_text`, `explore_answer`, and the reference `end_point` of implicit-intent tasks can leak the answer if passed to an evaluated model.
+
+## Evaluation protocol and answer visibility
+
+This is an openly released benchmark, so the Dataset Viewer and downloaded
+files include reference answers and evaluator geometry. The benchmark is not a
+hidden-label test set. Evaluation must construct the model-visible prompt from
+the task type and the live UrbanGround observation instead of passing an entire
+row to the agent.
+
+A conservative default allowlist is `id`, `task_type`, `task_abbreviation`,
+`capability_level`, and `description`. Task-specific controller code may also
+consume the appropriate start state or instruction parameters, but must keep
+answer fields, reference goals, route targets, restricted-zone geometry, and
+annotator metadata on the evaluator side. See the UrbanGround evaluation code
+for the authoritative per-task protocol.
+
+## Using the data with UrbanGround
+
+The dataset contains task definitions, not rendered observations or trajectories. Running an episode requires an UrbanGround desktop build or the browser application. Extract `raw/urbanground-app-task-files-v1.0.0.zip` so that its `task/` directory sits beside the desktop application, or use the task files already included in an official release archive. The archive contains the 728 stored App files; the application and normalized `test` split materialize the 110 Level-5 condition views from their source tasks at runtime.
+
+The Hong Kong 3D Visualisation Map, 3D Pedestrian Network, Unity runtime, Cesium, and Microsoft Rocketbox assets are **not** redistributed in this dataset. Those components remain subject to their respective upstream terms.
+
+## Data considerations
+
+- Task coordinates refer to public locations in Hong Kong and use WGS84 longitude/latitude.
+- Some questions reproduce public-facing business names, signs, addresses, or telephone numbers visible in the city data.
+- Descriptions involving people or households are task scenarios; the dataset does not contain records about evaluation participants.
+- The dataset does not contain participant trajectories, account credentials, API keys, private user data, 3D Tiles, screenshots, or application binaries.
+- Geographic content can become outdated as streets, businesses, and access conditions change.
+- Successful use depends on the matching UrbanGround application and external map services.
+
+## License
+
+The original task annotations, schema, and dataset packaging are released under
+the [MIT License](LICENSE), matching the UrbanGround project. External map and
+runtime components are not included and are not covered by this license. Public
+geographic and storefront facts reproduced in individual tasks remain subject
+to any applicable source terms.
+
+## Rebuilding the normalized files
+
+From the UrbanGround source checkout, run:
+
+```bash
+python HuggingFaceDataset/urbanground-tasks/scripts/build_dataset.py \
+  --source task \
+  --output HuggingFaceDataset/urbanground-tasks
+```
+
+The script validates the expected subtask counts, materializes the Level-5
+views, writes both JSONL splits, creates a deterministic App-compatible ZIP,
+and regenerates its SHA-256 manifest.
+
+## Citation
+
+```bibtex
+@article{ju2026urbanground,
+  title   = {UrbanGround: From Local Perception to Spatial Agency in a Real-Scale City},
+  author  = {Ju, Tianjie and Wu, Zheng and Sun, Yueqing and Cui, Yuhan and Li, Bobo and Wu, Shengqiong and Cheng, Pengzhou and Zhao, Haodong and Wu, Zongru and Ma, Xinbei and Zhang, Doris and Li, Kunling and Lee, Mong-Li and Hsu, Wynne and Fei, Hao and Gu, Qi and Liu, Gongshen and Zhang, Zhuosheng},
+  journal = {arXiv preprint arXiv:2608.27456},
+  year    = {2026}
+}
+```