Upload Multimodal-Mind2Web batch 1

README.md

@@ -0,0 +1,118 @@
+---
+license: other
+tags:
+- cua-lite
+- gui
+- sft
+task_categories:
+- image-text-to-text
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path:
+    - "*/*/train*parquet"
+    - "*/*/train/*.parquet"
+    - "*/*/train/*/*.parquet"
+  - split: validation
+    path:
+    - "*/*/validation*parquet"
+    - "*/*/validation/*.parquet"
+    - "*/*/validation/*/*.parquet"
+- config_name: web.navigation
+  data_files:
+  - split: train
+    path:
+    - "web/navigation/train*parquet"
+    - "web/navigation/train/*.parquet"
+    - "web/navigation/train/*/*.parquet"
+  - split: validation
+    path:
+    - "web/navigation/validation*parquet"
+    - "web/navigation/validation/*.parquet"
+    - "web/navigation/validation/*/*.parquet"
+---
+
+# cua-lite/Multimodal-Mind2Web
+
+cua-lite preprocessed version of Multimodal-Mind2Web (osunlp/Multimodal-Mind2Web). The screenshot-augmented Mind2Web web-agent benchmark: real human-annotated multi-step navigation episodes across 100+ websites. Only the train split is processed (the test_task/test_website/test_domain benchmark holdouts are never trained on). Episodes containing a native <select> step are dropped (no native select tool in the cua-lite action space).
+
+## Origin
+
+- [https://huggingface.co/datasets/osunlp/Multimodal-Mind2Web](https://huggingface.co/datasets/osunlp/Multimodal-Mind2Web)
+
+## Load via `datasets`
+
+```python
+from datasets import load_dataset
+
+# entire dataset
+ds = load_dataset("cua-lite/Multimodal-Mind2Web")
+
+# just one (platform, task_type) cohort
+ds = load_dataset("cua-lite/Multimodal-Mind2Web", "web.navigation")
+```
+
+You can also filter by `metadata.platform` / `metadata.task_type` /
+`metadata.others.*` after loading; every row carries a rich `metadata`
+struct (see schema below).
+
+## Schema
+
+Each row has these columns:
+
+| column | type | notes |
+|---|---|---|
+| `images` | list[Image] | embedded PNG/JPEG bytes; HF viewer renders thumbnails |
+| `messages` | list[struct] | OpenAI-style turns with `role` + structured `content` |
+| `metadata` | struct | `{platform, task_type, extra_tools, valid_actions, others{...}}` |
+
+Coordinate values in `messages` are normalized to `[0, 1000]` integers.
+
+## Layout
+
+```
+<platform>/<task_type>/<split>/shard-NNNNN-of-NNNNN.parquet                  # single-variant cohort
+<platform>/<task_type>/<split>/<variant>/shard-NNNNN-of-NNNNN.parquet        # multi-variant cohort
+```
+
+- `platform` ∈ {desktop, mobile, web}
+- `task_type` ∈ {understanding, grounding.action, grounding.point, grounding.bbox, navigation} — used verbatim as the dir component
+- HF config names are `<platform>.<task_type>` (e.g. `mobile.grounding.action`). The agent registry lookup key in code is `<agent>@<platform>@<task_type>` (e.g. `qwen3_vl@mobile@grounding.action`); only this user-facing token uses `.` between platform and task_type, because `@` triggers a 403 on the dataset-viewer's signed image URLs.
+- HF split names stay `train` / `validation` (the `datasets` library blacklists `<>:/\|?*` in split names; everything else is fine in config_name)
+- `validation` is an in-distribution held-out slice (never used in training); `test` is reserved for out-of-distribution benchmark datasets
+
+## Stats
+
+| platform | task_type | variant | train | validation |
+|---|---|---|---:|---:|
+| web | navigation | navigation | 583 | 19 |
+
+## Local mirror & SFT export
+
+For local workflows (SFT export, dedup, mixing across datasets), use
+`lite.data.hf.download` to mirror this repo back to the canonical local
+layout:
+
+```
+$CUA_LITE_DATASETS_ROOT/cua-lite/Multimodal-Mind2Web/
+  images/<hash[:2]>/<hash>.<ext>                          # content-addressed image store
+  <platform>/<task_type>/<split>[/<variant>].parquet      # rows reference images by relative path
+```
+
+Rows in the local parquet have `images: list[str]`; bytes are extracted to
+the image store. `lite.train.utils.data.export_sft` consumes the local
+form directly with `--image-root=$CUA_LITE_DATASETS_ROOT`.
+
+- Total unique images: **3,849**
+- Image store size: **3.71 GB**
+
+## Notes
+
+Train split only; test_* splits are benchmark holdouts and are not produced. Coordinates are the centre of the target element's bounding box, normalized to [0,1000].
+
+## License & citation
+
+See original dataset (osunlp/Multimodal-Mind2Web).
+
+See https://huggingface.co/datasets/osunlp/Multimodal-Mind2Web

stats.json

@@ -0,0 +1,11 @@
+{
+  "rows_in": 0,
+  "rows_out": 602,
+  "rows_dropped": 0,
+  "unique_images": 3849,
+  "image_store_bytes": 3709625117,
+  "by_partition": {
+    "web::navigation::train::navigation": 583,
+    "web::navigation::validation::navigation": 19
+  }
+}

web/navigation/train/shard-00000-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ed7a04ee3a934af9b06b442376df943244742d7adbb800266a8174b90ee4df8
+size 214411304

web/navigation/train/shard-00001-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7f8ab2a69aaa7fd1187cac01f4a55c4bb6d3373241cad96f7d2430a9138f05a
+size 171510811

web/navigation/train/shard-00002-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e80f60a75e6c2c54c7af6a99fa49327c9b04f795fb202caa88853c2ac819e295
+size 205109891

web/navigation/train/shard-00003-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90eface8df5425c9b9c6df23345bd616a6713435192179249f0c5061df2623dc
+size 164106826

web/navigation/train/shard-00004-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85b15419c30e5b0b96070b57640e2359d00eb0982101cf5e17c03a40c8af5f06
+size 182739137

web/navigation/train/shard-00005-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d966afe583c2e3215a0fde0b1ac42be479477f9cdc9a2501093bb1b162cf4be
+size 168675189

web/navigation/train/shard-00006-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5dca118e1db8fa883c2a344558ae29d9721f9326bc9329209b3b3fad9ba196b1
+size 170916992

web/navigation/train/shard-00007-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57089c5d3dc54835d8c68455914d125bf8fd9efeefe536d4da237995af3410d2
+size 200905586

web/navigation/train/shard-00008-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16928738e59f552d17231a0e549f46686d44c3e95d9ee788cb1801e2bba9a03f
+size 194933086

web/navigation/train/shard-00009-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03a835b4efc9cd345de41b0348eff471335257ff49261213cb612379af520623
+size 167432664

web/navigation/train/shard-00010-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e1e1960520253256825a52cc9b49fb7b6d01aa84bcb7ac1bfbdee0e99bb2788
+size 166112095

web/navigation/train/shard-00011-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86db174fb60dc7258fd3553741be1b7219853c5e41720fd3a91cb4d62f34a0d6
+size 159755602

web/navigation/train/shard-00012-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec5a6421b11ac25956767989a916b0d126ed1a5d08d56f999989e41a32c26501
+size 197984970

web/navigation/train/shard-00013-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:969195095462b1c54f9336408c2dcebb2fb79116c9aed694405460daa5daddf0
+size 189307036

web/navigation/train/shard-00014-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30cda02603e9b58c37f87e6caa14c16b8dd11bf8d69526a63dae7a05a96193ff
+size 187791018

web/navigation/train/shard-00015-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cef92941d3f17178521f74f1111bd34bb54ca3c43c7427a6f54715051d967fbf
+size 190274597

web/navigation/train/shard-00016-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:562a5edac95c1f95b8644d563cc1af4665a73d4c354386b2a42aba63c153366f
+size 176707553

web/navigation/train/shard-00017-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a7b46c4171d2ce900a2a2de32062038b43f1bca3181419f86ea1e32ee31e1f3
+size 227387643

web/navigation/train/shard-00018-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b33360e745698c3d6a71a49a1c5f1e9d3024b527ce9d6dadac8192281d0289a
+size 169146876

web/navigation/train/shard-00019-of-00020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a84ab21e3e2635fd8b1230e3f6ed085d98c93d3228fee36158a17686242dadf
+size 175277956

web/navigation/validation/shard-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b2d5a7d9f070d79bc340ef6ba6860a5539dfa4a74b5f2ba0a7b0a1686b1ddc2
+size 131958080