Upload GUIOdyssey batch 13

README.md

@@ -47,7 +47,7 @@ configs:
 
 # cua-lite/GUIOdyssey
 
-cua-lite preprocessed version of GUIOdyssey (hflqf88888/GUIOdyssey). A long-horizon cross-app Android mobile dataset of 8,334 task trajectories over ~128k screenshots. Produces two cohorts: navigation (multi-step agent episodes) and understanding (per-step screen captioning from the source `description` annotations).
+cua-lite preprocessed version of GUIOdyssey (hflqf88888/GUIOdyssey). A long-horizon cross-app Android mobile dataset of 8,334 task trajectories over ~128k screenshots. Produces two cohorts: use (multi-step agent episodes) and understanding (per-step screen captioning from the source `description` annotations).
 
 ## Origin
 
@@ -65,45 +65,114 @@ ds = load_dataset("cua-lite/GUIOdyssey")
 ds = load_dataset("cua-lite/GUIOdyssey", "mobile.understanding")
 ```
 
-You can also filter by `metadata.platform` / `metadata.task_type` /
-`metadata.others.*` after loading; every row carries a rich `metadata`
-struct (see schema below).
+After loading, parse `metadata` as JSON before filtering by `metadata_kind`,
+`dims`, or `others.*`; every row carries a rich metadata object inside that JSON
+string (see schema below). CUA rows use `dims == [platform, task_type]`.
 
 ## Schema
 
-Each row has these columns:
+Published parquet 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_tool_schemas, valid_actions, others{...}}` |
+| `messages` | string (JSON array) | parse as JSON to OpenAI-style turns with `role`, structured `content`, nested `tool_calls`, and `role:"tool"` results |
+| `metadata` | string (JSON object) | parse as JSON to fields `metadata_kind`, `dims`, `extra_tool_schemas`, CUA-only `valid_actions`, and `others` |
+| `_folded` | string (JSON array, optional) | folded grounding/understanding rows only; authoritative per-instruction `messages` / `metadata` members |
 
 Coordinate values in `messages` are normalized to `[0, 1000]` integers.
+The JSON examples below show the decoded shape, not the raw string cell.
+
+`metadata.extra_tool_schemas[*]` uses the nested Chat Completions function-tool
+declaration shape:
+
+```json
+{
+  "metadata_kind": "cua",
+  "dims": ["desktop", "use"],
+  "extra_tool_schemas": [
+    {
+      "type": "function",
+      "function": {
+        "name": "bash",
+        "description": "Run a shell command.",
+        "parameters": {
+          "type": "object",
+          "properties": {"cmd": {"type": "string"}},
+          "required": ["cmd"]
+        }
+      }
+    }
+  ],
+  "valid_actions": ["click", "type"],
+  "others": {}
+}
+```
+
+`messages[].tool_calls[*]` uses the matching nested invocation shape. Tool
+results pair `tool_calls[].id` with `role:"tool"` `tool_call_id`:
+
+```json
+[
+  {
+    "role": "user",
+    "content": [
+      {"type": "image", "index": 0},
+      {"type": "text", "text": "Click the OK button."}
+    ]
+  },
+  {
+    "role": "assistant",
+    "tool_calls": [
+      {
+        "id": "call_0000",
+        "type": "function",
+        "function": {
+          "name": "computer",
+          "arguments": {
+            "actions": [
+              {"action": "click", "coordinate": [640, 400]},
+              {"action": "type", "text": "hello"}
+            ]
+          }
+        }
+      }
+    ]
+  },
+  {
+    "role": "tool",
+    "tool_call_id": "call_0000",
+    "content": [
+      {"type": "image", "index": 1},
+      {"type": "text", "text": "clicked; typed"}
+    ]
+  }
+]
+```
 
 **Image-dedup (`grounding.*` / `understanding` cohorts).** These cohorts are
 single-image-per-row and many rows share the same screenshot, so to avoid
 re-embedding identical image bytes once per instruction they are stored
 *folded*: one row per unique screenshot (image embedded once), carrying an
 extra **`_folded`** column — a JSON string with the authoritative list of
-`{messages, metadata}` members for that screenshot. The row's top-level
-`messages` is the members concatenated for viewer convenience. `use`
-cohorts are not folded. **Use `lite.data.hf.download` to consume this repo** —
-it unfolds automatically back to one row per instruction; reading the parquet
-directly yields the folded form.
+per-instruction members for that screenshot. Each member's `messages` and
+`metadata` values are the same opaque JSON strings described above. The row's
+top-level `messages` is a JSON string containing the members concatenated for
+viewer convenience. `use` cohorts are not folded. **Use
+`lite.data.hf.download` to consume this repo** — it unfolds automatically back
+to one row per instruction; reading the parquet directly yields the folded form.
 
 ## 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>/<task_type>/<split>/<variant>/shard-NNNNN-of-NNNNN.parquet
 ```
 
-- `platform` ∈ {desktop, mobile, web}
+- `platform` ∈ {desktop, browser, mobile}
 - `task_type` ∈ {understanding, grounding.action, grounding.point, grounding.bbox, use} — 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 config names are `<platform>.<task_type>` by default (e.g. `mobile.grounding.action`) — UNLESS the dataset was staged with `--config-names`, which sets verbatim, explicitly-chosen config names (see the `configs:` block above for the authoritative list). 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
+- `validation` is an in-distribution held-out slice: no validation **sample** also appears in `train` — content-identical rows (same `images` + same `messages`, differing only in their ids) are co-located into one split, so upstream re-publishing one sample under two ids cannot straddle the split. It is *not* disjoint in **images**: one screenshot legitimately backs many distinct samples, and only whole samples are co-located, so the same picture can appear on both sides. `test` is reserved for out-of-distribution benchmark datasets
 
 ## Stats
 
@@ -135,6 +204,8 @@ form directly with `--image-root=$CUA_LITE_DATASETS_ROOT`.
 
 Source `info` coordinates are already normalized to [0, 1000] (per the upstream README) and are used directly — they are NOT device pixels.
 
+split: content-identical rows co-located, then hash_split on metadata.others.id with val_frac=0.02, seed=42, val_cap=2000 per (metadata.dims[0], metadata.dims[1], variant) (an upstream split label, where the source ships one, wins over the hash). A cap-bound carve depends on source iteration order and these parameters do not reproduce it; content co-location may make the final physical validation row count differ from the cap.
+
 ## License & citation
 
 See original dataset (hflqf88888/GUIOdyssey).

mobile/use/train/use/shard-00574-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65e8a425f91cf3da25718c79e9dad9a6f7d80cfb6b172c2b82f360f4f9d2763d
+size 115689472

mobile/use/train/use/shard-00575-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e03f6e197f93a18bbfb112e8f27e293bee0b93c28d27fa2da4ee677f3a833e18
+size 131477713

mobile/use/train/use/shard-00576-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5957925c3059b4769ff280febd5a5d9677cb30b6832dc47e263eadda5fa56264
+size 153341945

mobile/use/train/use/shard-00577-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6043c21649f61acc1b920c58c9de061d801cda8ecf0f8788c56d4fdf815c4379
+size 145496736

mobile/use/train/use/shard-00578-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fd3a4392ae75d210d74c7509e0ce9e3750d311ac8cdb835a7f819fc51816092
+size 142455766

mobile/use/train/use/shard-00579-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e51f33b7daddda5a791aaf7e4f06b32c78007c554fdb86818a25a75e1763a90
+size 140325471

mobile/use/train/use/shard-00580-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82944073c3c99c3a00b1cc3f4f072089e97b0f72a0d5686b25c2f26916a16c40
+size 139038386

mobile/use/train/use/shard-00581-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dad06f2eb046205bf9a58c22bef05e6610ea10a18c68ba5bceee111cbed23592
+size 114679901

mobile/use/train/use/shard-00582-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2d5daa32dd93e15e0fd4dc7f919f0f6a983e7911ce67c5fa8e0186bd7670681
+size 168463669

mobile/use/train/use/shard-00583-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d99cac1e5d5de97834895fed1dd6e24940ed0c4ea039a76d190e84edf490070f
+size 138565267

mobile/use/train/use/shard-00584-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9270238d43ab137c9d0a498cc05f7e57f1acdc8129d44da30244b3feaacd464
+size 186239938

mobile/use/train/use/shard-00585-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c597d326eeb570dce11ab9f84925a2d7d9263e9376ba9ec637c63bfc01242dd1
+size 141478854

mobile/use/train/use/shard-00586-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:975f4eafb051d3a0eb2a1f1d95db38ed9c172d1d2e175941d9880a6eb3e9ea34
+size 144116304

mobile/use/train/use/shard-00587-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1f344d7971c305a939e1440dd17a6c9a09d308b1b3ff2db8ecaed552990bc9
+size 158757849

mobile/use/train/use/shard-00588-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83fa9e22ea1de195e074a031e3adc624ddabd5671b298fceba03c20429ab9444
+size 124471548

mobile/use/train/use/shard-00589-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94c500d10d7d450ff8591a4ceef91c7ff946bb75ebea4a154b8beb6da82ebfc8
+size 148932524

mobile/use/train/use/shard-00590-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62bf01ba27e606a1577f323243c7dca6d45f3695435d637b73b6d992b9a5cebe
+size 114233594

mobile/use/train/use/shard-00591-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d67d2a758b176a84353b934698dc36a805ab4d7395096e1012b0d398f237105f
+size 121301414

mobile/use/train/use/shard-00592-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b91494c98f6772e85bca2811ca0ea6cccd789e4c50bfd55d293f03376db8048
+size 114957558

mobile/use/train/use/shard-00593-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4081dc9d61649de0edf88d6a0ba7e7e1e8515bf73a4f3ac6847ee989040c559
+size 151244446

mobile/use/train/use/shard-00594-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3a7b18dc6ea163eaf9d3d7104d8d7a6fb2d363ff013bd4539f3e630ec8e8aa8
+size 114600014

mobile/use/train/use/shard-00595-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ed45a6b9eabdbfd6da35d2d8f0274e1c86db0dd1eb9219266b5284f09395851
+size 150041505

mobile/use/train/use/shard-00596-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d22f64b143879480de07ec7dc1df157971d052621e37c255960edcc323fd060
+size 129986432

mobile/use/train/use/shard-00597-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:767a9b2ad3868c78add861bcd334ed80592e1431e6a2016120df1dbca1f249e1
+size 154351808

mobile/use/train/use/shard-00598-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ce0bd81fbf942c5503023739148c085b945633c884d24e492fd97fb9b6ec752
+size 121730351

mobile/use/train/use/shard-00599-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1140fe8a8274f79e6291a3e559bf563b97b19e3a1a25edb8ae4364bad8c71c19
+size 118044672

mobile/use/train/use/shard-00600-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:badb2f2efe6328bdcb72df5df90c8b197f87281172ab4b4d1b43ee3106d0d220
+size 109727966

mobile/use/train/use/shard-00601-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a75ba1f795d5735d98c378e13bd5201b377a94e3693cbb37a47c3cecc298612e
+size 123938158

mobile/use/train/use/shard-00602-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04d3c0d2e2df418336f37a32dd1a5e7a76f7e315834009750919e1150e7e72fd
+size 153152324

mobile/use/train/use/shard-00603-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:401a40bec902c315936bc6a8e46dc727001be5a09640f226883fbdb22a028ed4
+size 142988735

mobile/use/train/use/shard-00604-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e674281873200e18b36c2f6efe07fbabe9cea6665f43ae01c65646c8b34589ff
+size 142883160

mobile/use/train/use/shard-00605-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5dc3e5fb7776a1abe1dfaaeb2b7eb8ff9c41adc15d4c2292a901b20acad3e44
+size 137396901

mobile/use/train/use/shard-00606-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd104dc9d55c2d678308d49eeaa7c77bfaca9191e5ec3aed57ef09abc1a69989
+size 156781547

mobile/use/train/use/shard-00607-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa261922f67e0fb6eb880effabc0a6ba617d1aa63daa2281367b8dd24385a297
+size 149074899

mobile/use/train/use/shard-00608-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd39f177659087342b10ed325dcb8fd03223f06595746b4853fbe91d11d7cf07
+size 142422136

mobile/use/train/use/shard-00609-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11f5e625bd4ac82fdf9e51368ff1cc88d9ae863f74b0623b0fa860a264b78c1a
+size 148894691

mobile/use/train/use/shard-00610-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45ee28ee4a206523c1c3d5840b7f9538c114ee8b9d5eff54223ce8aa48cd89fa
+size 132503093

mobile/use/train/use/shard-00611-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a04da038c0158cbfb5252e639306f49cf7a49fc67af89ae8c3a7173c0521ea8
+size 148524918

mobile/use/train/use/shard-00612-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14314fe06633809cf5d6d40f046a3c74feeb7149d7a67155030a6a1ded6aa70c
+size 112784904

mobile/use/train/use/shard-00613-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7c3d2d960c5de4e75b15bdfb76f78b6795728c4428d1ea137f47104f3da4a57
+size 160724704

mobile/use/train/use/shard-00614-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7da7d7c364fbb5f79e4c887b5cd990705933de0c73a3efa3ca3d5ef33644305
+size 165968742

mobile/use/train/use/shard-00615-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ba8d6ace1d586885d74fdf52ecfd5038745127c714b5c170317ea63f79f6515
+size 121254129

mobile/use/train/use/shard-00616-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f864d5018a7f2f274b52e58475a8dfb874e6ad7276ba52c491bd6d19877e54af
+size 118241357

mobile/use/train/use/shard-00617-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62d117cd6ea4181e0cb0959687527cb3a932af615635dac66e0bcc7d4cd5bdc0
+size 95344294

mobile/use/train/use/shard-00618-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6c58b94c0a46bc671df65806c7ae67cdffd19c24df7891e9ea0629f6122c382
+size 140574986

mobile/use/train/use/shard-00619-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28391a1506c9641cf968c6bf407bea2ac064f76dafd34120804cf6c3f5b30de7
+size 107969309

mobile/use/train/use/shard-00620-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66de37935607a6e88938298c3fe903566c09db7332886bc29f2699ffa9876716
+size 138611517

mobile/use/train/use/shard-00621-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40cafcba0f083e599d826ae8c702683b7d66aa5498098634895304f3c36dc880
+size 134836012

mobile/use/train/use/shard-00622-of-00652.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca2f114de0817337e9580fc4ce166b7987b381c886f26bd0d5bd74d188ff6702
+size 122359097