Glance evaluation
Collection
Public evaluation sets and full evaluation runs for Glance. • 2 items • Updated
Error code: DatasetGenerationError
Exception: CastError
Message: Couldn't cast
vid: string
duration: double
n_frames_1fps: int64
n_frames_2fps: int64
qid: int64
text_sha1: string
to
{'qid': Value('int64'), 'vid': Value('string'), 'text_sha1': Value('string')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1835, in _prepare_split_single
for key, table in generator:
^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 621, in wrapped
for item in generator(*args, **kwargs):
~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 373, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2394, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2322, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
vid: string
duration: double
n_frames_1fps: int64
n_frames_2fps: int64
qid: int64
text_sha1: string
to
{'qid': Value('int64'), 'vid': Value('string'), 'text_sha1': Value('string')}
because column names don't match
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1379, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
~~~~~~~~~~~~~~~~~~~~~~~~~^
builder, max_dataset_size_bytes=max_dataset_size_bytes
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 956, in stream_convert_to_parquet
builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1702, in _prepare_split
for job_id, done, content in self._prepare_split_single(
~~~~~~~~~~~~~~~~~~~~~~~~~~^
gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
):
^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1887, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
qid int64 | vid string | text_sha1 string |
|---|---|---|
0 | 3MSZA | 1b112fe129327b4d3e32b48a2dfed2792ac31afa |
1 | 3MSZA | 75e51b1ec5848caf6a33b85e4a51bb0640e99992 |
2 | 3MSZA | 84a7b766fb2fe5e5395b19e1ffcc5d85c3965c53 |
3 | 3MSZA | cbd1833214a0331f720906fe4472c73f183e550e |
4 | AMT7R | 66994de6c5c59f2751cbb10acad40695223fa3c2 |
5 | YVKIV | 572a52f6c574ca16c2676786d5063196d7b6e9cb |
6 | VXJS4 | f5b24e63bbde81f4356c37aa2360d445edf961ca |
7 | VXJS4 | 5c4c5799ed6b69f53f09222a64b6e04125018df0 |
8 | GBD1Y | d1c163953d7b63b2b128cbbbeb1192971c15ed3f |
9 | GBD1Y | fa042daf32990d1947f2c13b35db1d88286db7a6 |
10 | KVXJ9 | cd4de9b5b75e4453bf7ca26332b377f51977cedb |
11 | KVXJ9 | 2bd52f8cd74c453dab080d5f4b7ad379e19553db |
12 | E6DLK | 6c6e8e3b8f487447df95213fbd50f2cc5b1e0c4b |
13 | E6DLK | 248cf37f06e28874ec0e6e65b3093ee14e90df58 |
14 | E6DLK | 7e3e7f18bda3d9aa2b853d2ae7abaae0ee1e42ed |
15 | E6DLK | b19431e27b8cd8b1fefed9251df2e75bce130a11 |
16 | E6DLK | 3ce548310641ef91ef7de95f1c6c883d8597b68f |
17 | E6DLK | 2fa1c1f845da84058a8c06801e93d3637f1b1157 |
18 | AKO6M | 51f2f01391c91a49f16af00d6b612f472ece9e9e |
19 | AKO6M | 19c270c32b0305be3eda1bcc5a5aaa2f584cd13e |
20 | F7TG5 | f0909b0c4ac6433b6c2b326f2854fb0f06dc51dc |
21 | F7TG5 | 683a5920e655b0432c63593fab36244ba975ed77 |
22 | F7TG5 | 967d512b42addf5ad58e550cccfc282eaab49191 |
23 | F7TG5 | 08f88d8e45177ee5a5655256b9221207000b42c1 |
24 | KOVTR | 03ae23ff5da978bfdbdd969c026327abf5d117d4 |
25 | KOVTR | 8ed601dd4d46a752f9d20263c4118a1c5ba3b301 |
26 | KOVTR | ce190e5805c96089139e8ca45705a287c761137f |
27 | FPJ9D | d129e59860f2cd12a5c6bfa66ef24773f7099289 |
28 | 5NV2Z | 61ae775dcc61e1a94c151c2a9e6a579581298c00 |
29 | 5NV2Z | b81e0d3a60c402c1590262812b63ab42f6c26699 |
30 | 5NV2Z | 6324f134d111cd544a9d2d1f6bfae8297b33b1ac |
31 | 5NV2Z | 272fb9ae5f1b08836f9806f806529da1d38ad8ca |
32 | Y1HGC | f12003354a17ff70ef537b5bc80eabac7e9e57bb |
33 | Y1HGC | 83c8938851787a0c3d07c6afb01690b8646b1720 |
34 | 9JZO2 | 706713d508747b3037fc5b97d682191ce733bb95 |
35 | 9JZO2 | 79e8973279b56dc32d2733549f642da678cf6215 |
36 | 9JZO2 | 0d7ba4449a6af876350d637d522646735343694b |
37 | 9JZO2 | a363e5d67de1ff6798c7c5d5605a1b9d448151a0 |
38 | 9JZO2 | ee5274375af5ff4f3a3e0d58b0a515cc6090d0a9 |
39 | 9JZO2 | 612b02c134364a2a85e21141890335e6a8e84de9 |
40 | 9JZO2 | da57552df23ced6ce40ef8dab108d8b56b9da822 |
41 | 9JZO2 | 11b5ee95b4dccdb9b38a5f32c3681d540e024685 |
42 | 9JZO2 | 23ecfdaf5eefbff73d42a14eeb6c267b4b12bf00 |
43 | 9JZO2 | 76353ba8b9f1322f50fab8371e19b3a58a8f94e3 |
44 | 9JZO2 | 77ce3f402a99a459c6e7fa4d60910adb9f9299ba |
45 | 9JZO2 | 783c24d72b4d37389eb382c777e5e4cb7b35c32a |
46 | 4J1AP | 83c8938851787a0c3d07c6afb01690b8646b1720 |
47 | 4J1AP | 276d9249f30c8efd3730dd8ba50a657058d29a3d |
48 | V1WN7 | 05e11374ae821421d8054c64ea094c3f81f4a779 |
49 | V1WN7 | 511bd7e3b2d30a81807d857bbc73d2e15bb5251a |
50 | V1WN7 | eef6c7f20591a9b7a420a47bcdbfe352c5c3d8f1 |
51 | BL7OF | bc7affeb5eb7eb6e9cfa2a59b4b5dc4631ae946e |
52 | BL7OF | 01bf939cc63323b5046cd027d3a5fe597dae40e3 |
53 | 5C4EK | 1d1f6337adfb61a7534dd8a44a2b209b847efd19 |
54 | S591U | baa099590a7717ed29e24ff06c1657913f69f4cc |
55 | DLFPX | eceabc894faae1233c667c5b6b016fed171bb63f |
56 | 9HG4Z | 8682d2ee3084f8fe281fb344585a5d6ac0920995 |
57 | I7AS7 | 844d7dc57eeb676699c4dfaf7bf35de2f18d1267 |
58 | I7AS7 | 83cab41370159b528082fef0eeabd027b798a6ad |
59 | I7AS7 | afcfb968c5ba0ab261f057f83f7e16c2d11b5bd6 |
60 | J48N6 | c9a854a96303287c746935431014200422fc9dbd |
61 | J48N6 | 6fa08457f12585979cd5953e261b9d8608dd0381 |
62 | J48N6 | a67dcd50becc7338e619702d3d421f9c83c2c868 |
63 | JVLAZ | f4c530372007e0f516e99e652352e94745d220cf |
64 | JVLAZ | aa6cc354c1cb77b3c0afba5616d59d8267af843f |
65 | JVLAZ | 900d788ad89bfccc9d2514831d47d2ead9cc9f91 |
66 | 2RFLZ | 7b153cb79770d68588b1d3896e2db2d8903b3956 |
67 | 2RFLZ | e6644da2fb8736ae87cba238aa2782fa7d1c4171 |
68 | N3U9S | 1b0d391d61d4bd053a530cb5bf48b1a790526798 |
69 | J9T5D | 4704d1268346a726b2ce663f93cc953964f62693 |
70 | 1CYLM | 161fdba58a0fe002a670bc1f63a8806bbb506663 |
71 | 1CYLM | 80f0eb83f49b9b4e6cd02cd32158c8ddecddb658 |
72 | V2GC9 | 4530d0fd2a10406b65dc43518a05fcc7758c1b4b |
73 | V2GC9 | ed0586ff925b3a9777f931fa33e45e8ee0eaaa0e |
74 | V2GC9 | 698c8d9302ef47c4921237303763f0c98c9e5588 |
75 | V2GC9 | 28aea6acf9b49afdfd6941f83067fa81454a7059 |
76 | LGJAR | 744548aea6c498c46b606181ec317291abe9ca9a |
77 | ZHRPD | 1f3589bae38691289bfda3b8c6b9cecd2ff37c7a |
78 | ZHRPD | 3883d294e55c13e8cc5e0d8db8733139b38748c4 |
79 | 43CCM | be89342313f62c93e78cffd7b8f42dfa9b73624c |
80 | 43CCM | ee0c4faff264d63fc4cee60c9442e75105668f87 |
81 | YZ8HK | d1cf12dc61491c8407860d32c2b9b25e41f2dbee |
82 | 2P7A9 | 7756c7a7a147ec9fadaea54cf332872e927785c7 |
83 | 2P7A9 | 0f0cad00a548267665649f5db08b94775965cdb7 |
84 | 2P7A9 | 1b3435ecd1e6c8cc327d72b4a94d6ca96c96a2e0 |
85 | 2P7A9 | 01912dd5afa8c11055584032417d71431a2f17a2 |
86 | 2P7A9 | 4c43a0c16e5f4a402b213e1fb5842268b8b321da |
87 | 2P7A9 | 31ed9cba6dded133ad488027651f980b2e79effa |
88 | 1ZWPP | d8e7a42d88fcefaed15a77c7eaa90ba31d325a26 |
89 | 1ZWPP | f6377022bcd7590e3e3b65dd6acd965f8db18c0a |
90 | 1ZWPP | ab1e95da8cbd5b75764eabd01228105cd6dc3f74 |
91 | QUXYH | 3878a18b7d0185dba446fa8055250343c8a54316 |
92 | QUXYH | 96a29e3f47dc2d2a328c39c662954983d9c35fb1 |
93 | QUXYH | 268dcd61157fa41d472760acf3c02030e4361965 |
94 | QUXYH | d80f0c65cb179fac976671803cfdcdd7b5af3154 |
95 | QUXYH | 9db3dd2031f104b288826079e64dc8871106aa8c |
96 | QUXYH | 68d1e4e3d59945ccd57ca58ed58b52d76faaac89 |
97 | V3SOF | f2261d91629fb3fc02ff79472ba03e846bd4963b |
98 | SY5QP | 7cdb6b3851137bcbf180ffdd0804b66362dc7806 |
99 | SY5QP | 1c3b463417c073fb0ba4d0fc550dc79c3783ff51 |
Results, model outputs and probe scores behind the Glance papers and the Glance engine. Nothing here is training
data. Each folder was published as its own repository before; its original card is the folder's README.md
(only links to the old repository updated), with the full provenance, the list of what was removed and why, and the scan method.
| folder | contents | formerly | licence |
|---|---|---|---|
grounding-negatives/ |
how much of a "no target" answer can be read from the sentence alone, across nine sources; per-item outputs of 14 grounding systems. Text replaced by SHA-1 | grounding-negatives-clean | CC BY-NC-SA 4.0 |
decision-index-v1/ |
Glance-4B-preview (v1, b6237b4) on the Decision Index public suites 0.3: text 41.26, vision 55.46; payloads removed |
di-runs-clean | CC BY 4.0 |
scan-engine/results/ |
every structured result of the scan-engine series: exactness, latency against vLLM/SGLang, memory tiers, retrieval quality, Visual Haystacks, absence detection, agent runs | collection-scan-results-clean | CC BY-NC-SA 4.0 |
scan-engine/data/ |
annotations and scores for the Charades hard-negative set and the agent experiment | collection-scan-data-clean | CC BY-NC-SA 4.0 |
scan-engine/charades-pilot/ |
Charades ids, query SHA-1s and per-clip / per-frame scores of the scan pilot | scan-pilot-annotations | CC BY-NC 4.0 |
architecture/ |
aggregate results of architecture experiments: readout α sweeps, GDN ablations, zoom and multi-head variants (RefCOCO testA/B, gRefCOCO val, ScreenSpot-Pro) | glance-train-data-clean arch/ |
CC BY-NC-SA 4.0 |
glance-coco-diagnostic-predictions/ |
Glance v1 and v1.1 predictions on the Glance-COCO diagnostic split, one row per request |
glance-ground-v2 eval/ |
CC BY-NC-SA 4.0 |
No video, frame, image or third-party benchmark text is included. Rows carry the source dataset's own ids, or the SHA-1 of the text, so they can be joined to your own copy.
Related data: Glance-COCO (image grounding), Glance-GUI, Glance-Video.