diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..a363f6ecfbe8e11faf052d881587b834fbde1851 --- /dev/null +++ b/README.md @@ -0,0 +1,91 @@ +--- +license: cc-by-4.0 +task_categories: +- visual-question-answering +language: +- en +pretty_name: GameBind counterfactual twins +size_categories: +- 10K/` with: + +- `clips.json`: one entry per clip, with the base scene, the twin, the events, the agents and the paths of its frames; +- `queries.json`: the questions asked on each clip, with their options and the correct answer; +- `checks.json`: the pixel checks of every twin; +- `summary.json`: the generator settings and the check summary; +- `frames/`: the images. + +The frame paths in `clips.json` are relative to the folder where you unzip. A GameBind clip has 8 frames +(`f0.png` to `f7.png`, 512x512 pixels), and the models in the paper saw the last frame, `f7.png`. A colour-shape clip +is one 512x512 image. + +## Twins + +| twin | change | +|---|---| +| `S` | base scene | +| `minus_c` | the competitor stays idle at its start tile | +| `minus_cue` | the target moves as in S, but its action leaves no trace | +| `minus_cue_minus_c` | both changes | +| `minus_ccue` | only the competitor's trace is deleted, the competitor still moves | +| `degraded_cue` | the target's trace is weakened | +| `swap` | the two agents exchange their events | +| `placebo` | an irrelevant change to the floor pixels | +| `S_repro` | a fresh rendering of S, used to check that rendering is exact | + +In the colour-shape displays, the twins grey out the competitor shape (`minus_c`), the target shape (`minus_cue`) or +both, add a grey shape elsewhere (`placebo`), or exchange the two colours (`swap`). A blind grey image (`blind`) is +also included. + +## Questions + +On every GameBind clip three questions are asked, each with four options. The joint question asks what the target +agent did, and its options are the true event, two answers that mix the target's and the competitor's action and +object, and the competitor's whole event. The object question and the action question ask for one part. Options are +shown under rotated orders, and all twins of a base scene share the same orders. The colour-shape question asks for +the colour of the target shape, with the target's colour, the competitor's colour and two absent colours as options. + +## Model answers + +`per_scene/` holds one gzip CSV per model and scene set: 19 open models on all scenes, and GPT-4.1, GPT-4o, GPT-5.5 +and seven open models on the API subset. Each row is one question about one twin of one base scene, with the answer +probability of every option. The subfolders `colorshape/` and `photo_twins/` hold +the answers on the colour-shape displays and on the real-photo twins (numbers only). `per_scene/README.md` describes +the columns. + +## Notes + +- All images are rendered. The dataset contains no photographs and no personal data. +- The renderer regenerates every frame bit for bit. The code is linked from the paper. +- The API subset is a separate draw from the 1,000 scenes, although the scene names repeat. A scene is identified by + its set and its `base_id`. + +## Licence + +CC BY 4.0. diff --git a/colorshape_twins.zip b/colorshape_twins.zip new file mode 100644 index 0000000000000000000000000000000000000000..67522fb2ee56d118df04923ce56501a89d7ba476 --- /dev/null +++ b/colorshape_twins.zip @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:33d9ca57f6090d5395f33c41126e2daf001c5d1e6e0bdb132b613dae841e68b9 +size 18174964 diff --git a/colorshape_twins_hard.zip b/colorshape_twins_hard.zip new file mode 100644 index 0000000000000000000000000000000000000000..0b776ef8f5c25cf063816effd28b280230e25b9e --- /dev/null +++ b/colorshape_twins_hard.zip @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b07cf497bbf0a2f5dee3369b1b386a7814b3a80b506da4729a6d8237ec1044bb +size 27647293 diff --git a/gamebind_twins.zip b/gamebind_twins.zip new file mode 100644 index 0000000000000000000000000000000000000000..a7f03232f67769d5edc4bbaac6e91853858778fb --- /dev/null +++ b/gamebind_twins.zip @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:92ee9813dc4638a9477ae04030dfb49c020326d494e26146571fe72009e3d4a4 +size 90380758 diff --git a/gamebind_twins_ext.zip b/gamebind_twins_ext.zip new file mode 100644 index 0000000000000000000000000000000000000000..0335bbb2ed8239b32e97f89af10490cea2cc7185 --- /dev/null +++ b/gamebind_twins_ext.zip @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f104cbbfa9edc08ed5414229c54d0f320cd04ac34abd7f58f037061a6dc5e477 +size 120573842 diff --git a/gamebind_twins_ext_api.zip b/gamebind_twins_ext_api.zip new file mode 100644 index 0000000000000000000000000000000000000000..f01212f88489ba41bc1dd2d91e12367fd968e5fd --- /dev/null +++ b/gamebind_twins_ext_api.zip @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0c363d94bef38febd1b8031885ff6c571db1b31f2e5e75698e61f38611bc7439 +size 35016804 diff --git a/gamebind_twins_full.zip b/gamebind_twins_full.zip new file mode 100644 index 0000000000000000000000000000000000000000..ea19a1b3e591adb56aef0dcfb2241b6ea61b885e --- /dev/null +++ b/gamebind_twins_full.zip @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ffa074d1a93c4491df0f2dcbe6dc59cb6eb99835b89e4816c127e8990d8aa1c5 +size 283697956 diff --git a/per_scene/README.md b/per_scene/README.md new file mode 100644 index 0000000000000000000000000000000000000000..65518c41491fff834739059a8ace7723173fff45 --- /dev/null +++ b/per_scene/README.md @@ -0,0 +1,73 @@ +# Per-scene rates + +One gzip CSV per model and scene set (`__.csv.gz`) and an index (`models.csv`). Every row is one +scored question about one twin of one base scene: model x base scene x twin x question type. These rows are +everything Table 2, the data figures, the appendix tables, `numbers.tex` (the numbers quoted in the text), the +sensitivity analysis, the heterogeneity, scene-level and extra checks are computed from (`reproduce/reproduce.py`). +They were exported from the raw scoring outputs by `code/scripts/export_per_scene.py` +(`export_per_scene_photo.py`, `export_per_scene_colorshape.py` for the two subdirectories). + +**Scene sets.** `full`: 1,000 main + 300 near base scenes (seed 424242). `api_subset`: 300 main + 150 near base +scenes (seed 31337), scored with the API models and seven open models. The API subset is a separate draw, not a subset of the 1,000 +scenes, although the base-scene names (`tw_main_g_0000`, ...) overlap: a scene is identified by +(`scene_set`, `base_id`), and every file holds exactly one scene set. + +| file | scene set | twins | rows | +|---|---|---|---| +| 19 open models (`qwen2.5-vl-7b-instruct__full.csv.gz`, ...) | `full` | main: S, minus_c, minus_cue, swap, placebo, minus_cue_minus_c, minus_ccue, degraded_cue, S_repro; near: S, minus_c | 23,040 each | +| `gpt-4.1__api_subset.csv.gz` | `api_subset` | main: S, minus_c, minus_cue, swap, placebo, minus_cue_minus_c, minus_ccue; near: S, minus_c | 6,486 | +| `gpt-4o__api_subset.csv.gz` | `api_subset` | as GPT-4.1 | 6,486 | +| `gpt-5.5__api_subset.csv.gz` | 250 of the 300 main API-subset scenes | S, minus_c (binding question only) | 500 | +| 7 open models on the API subset (Qwen2-VL-7B, Qwen2.5-VL-7B, Qwen3-VL-8B, Qwen3-VL-32B, Mistral-Small-3.1-24B, InternVL3.5-38B, Qwen2.5-VL-72B: `__api_subset.csv.gz`) | `api_subset` | as GPT-4.1 | 6,486 each | +| `photo_twins/*.csv.gz` | 463 real-photo target instances | S, minus_c_grey, minus_c_blur, placebo_grey, placebo_blur | 6,237 per model; blind run 1,389 | +| `colorshape/__{easy,hard}.csv.gz` | 600 colour-shape base scenes per display (5 open models) | S, minus_c, minus_cue, minus_cue_minus_c, placebo, swap, blind | 4,200 per model and display | + +## Columns + +| column | meaning | +|---|---| +| `model` | display name used in the paper | +| `scene_set` | `full` (`data/gamebind_twins_full` + `data/gamebind_twins_ext`) or `api_subset` (`data/gamebind_twins` + `data/gamebind_twins_ext_api`) | +| `condition` | `main` (the two events far apart) or `near` (the target's two candidate objects 1-2 tiles apart) | +| `base_id` | base scene S (e.g. `tw_main_g_0000`; `g`/`t` = grid / textured skin) | +| `twin` | `S` (base scene), `minus_c` (competitor idle), `minus_cue` (target's action leaves no trace), `swap` (events exchanged between agents), `placebo` (task-irrelevant floor pixels), `minus_cue_minus_c` (double deletion), `minus_ccue` (competitor's action leaves no trace), `degraded_cue` (target's trace at reduced strength), `S_repro` (an exact re-rendered copy of S, scored in the extension jobs: a scoring-calibration check) | +| `qtype` | `binding` ("What did the red agent do?"), `object`, `action` | +| `skin`, `target_agent` | skin of the scene; the queried (target) agent, `red` or `blue` | +| `target_action`, `target_object` | the target's event (a, o) in S | +| `competitor_action`, `competitor_object` | the competitor's event (a', o') in S | +| `primary` | 1 if neither agent's action is `go to`: the a priori primary subset of the processing tree | +| `answer` | canonical index of the correct option in this twin (empty where no answer is defined: `minus_cue`, `minus_cue_minus_c`; `swap`: the target now performs (a', o')) | +| `p0`..`p3` | answer probabilities in canonical option order, averaged over the option rotations (soft response). binding: [(a,o), (a,o'), (a',o), (a',o')]; object: [o, o', fill, fill]; action: [a, a', other, other] | +| `pred` | canonical index of the argmax of `p0..p3` (hard response) | +| `p_competitor` | P(answer names the competitor's action a'): `p2+p3` for binding, `p1` for action; for object P(answer names o') = `p1` | +| `correct` | 1 if `pred == answer` (empty where no answer is defined) | +| `n_rotations` | option rotations averaged: 4 (open models, next-token letter probabilities), 2 (GPT-4.1 and GPT-4o, top-20 log-probabilities of the answer letter; GPT-5.5, one parsed answer per rotation, so its soft response is a vote share) | + +Probabilities are written with full double precision so that the reproduction is bit-identical; the rows of each +model come from one scoring run (one GPU type, one batch size). The per-rotation probabilities are not released (only +their mean `p0..p3`); the analysis uses the mean and the argmax only. The photo-twin files have the same columns plus +`protocol`, `dataset` (vcoco / vg) and `photo_id`, without `scene_set`, `condition`, `skin` and `primary`. + +The colour-shape files (`colorshape/`, question "What color is the ?") have the columns `model`, +`display` (`easy`: `data/colorshape_twins`, large saturated shapes, 1-2 grey distractors; `hard`: +`data/colorshape_twins_hard`, small shapes among 14 coloured distractors, colours at 15% saturation), `base_id`, +`twin`, `distance` (near / far), `target_shape`, `competitor_shape`, `target_color`, `competitor_color`, `answer` +(empty for minus_cue, minus_cue_minus_c and blind), `p0..p3` (canonical order [target colour, competitor colour, +absent, absent], mean over 4 rotations), `pred`, `lm0..lm3` (probability mass on the four option letters before +renormalisation, per rotation) and `n_rotations`. + +## Example: GPT-4.1's competitor effect and processing tree + +```python +import pandas as pd +d = pd.read_csv("gpt-4.1__api_subset.csv.gz") +d = d[d.condition == "main"] +b = d[d.qtype == "binding"].pivot(index="base_id", columns="twin", values="p_competitor") +print("kappa_A =", (b.S - b.minus_c).mean()) # all 300 scenes (Table 2: 15.5 pts) +prim = d[(d.qtype == "binding") & (d.primary == 1)].pivot(index="base_id", columns="twin", values="p_competitor") +prim = prim.dropna(subset=["S", "minus_c", "minus_cue", "minus_cue_minus_c"]) +pi_S, pi_c, b1, b0 = (prim[t].mean() for t in ("S", "minus_c", "minus_cue", "minus_cue_minus_c")) +p_A = 1 - pi_c / b0 +m_A = (pi_S - (1 - p_A) * b1) / p_A +print(f"n={len(prim)} p_A={p_A:.2f} m_A={m_A:.3f}") # n=145 p_A=0.82 m_A=0.153 +``` diff --git a/per_scene/colorshape/gpt-4.1__hard.csv.gz b/per_scene/colorshape/gpt-4.1__hard.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..4eeda1891b650176157a47ca4a9be00bb2231e8f --- /dev/null +++ b/per_scene/colorshape/gpt-4.1__hard.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:670b58c5ecce1fef52f60a7bda987ff5e3fe11e53c13da7a5b62cae5426b4f93 +size 198798 diff --git a/per_scene/colorshape/internvl3_5-8b-hf__easy.csv.gz b/per_scene/colorshape/internvl3_5-8b-hf__easy.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..a0f1fd6f3f3dacd480b2277bf896a8fba7adb3b7 --- /dev/null +++ b/per_scene/colorshape/internvl3_5-8b-hf__easy.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be4855e41d87bc2084710353164bd85ae38a97761a927ce54cefe3ade5b372dd +size 308025 diff --git a/per_scene/colorshape/internvl3_5-8b-hf__hard.csv.gz b/per_scene/colorshape/internvl3_5-8b-hf__hard.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..61628ba6e309d4677c263153beaf404a73fdc779 --- /dev/null +++ b/per_scene/colorshape/internvl3_5-8b-hf__hard.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a44f58457025870237f52281a9fbcab63eac3f828d79be10d11d72e10e865ce +size 316102 diff --git a/per_scene/colorshape/llava-onevision-qwen2-7b-ov-hf__easy.csv.gz b/per_scene/colorshape/llava-onevision-qwen2-7b-ov-hf__easy.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..7a27547c25c432488977c218b6294ef6e8b7f4f4 --- /dev/null +++ b/per_scene/colorshape/llava-onevision-qwen2-7b-ov-hf__easy.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7b9e37a609be15d99c512854644f2e2c1cf7b8fe355cd7e28100e6e950e4069 +size 321101 diff --git a/per_scene/colorshape/llava-onevision-qwen2-7b-ov-hf__hard.csv.gz b/per_scene/colorshape/llava-onevision-qwen2-7b-ov-hf__hard.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..f42ccb4f94465d7db6fe8b7b8200ade2edd8e66c --- /dev/null +++ b/per_scene/colorshape/llava-onevision-qwen2-7b-ov-hf__hard.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31e8bf11405976e613a367d4a716606222eb629db8d109a619df2b9b4def255b +size 325584 diff --git a/per_scene/colorshape/qwen2.5-vl-7b-instruct__easy.csv.gz b/per_scene/colorshape/qwen2.5-vl-7b-instruct__easy.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..bb8be20993ade500377e4815c176e682f23fdfe3 --- /dev/null +++ b/per_scene/colorshape/qwen2.5-vl-7b-instruct__easy.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd6824b9d980a7739e2d5243bf0584ee4645e0b1c6840818420db36aeb563bab +size 321084 diff --git a/per_scene/colorshape/qwen2.5-vl-7b-instruct__hard.csv.gz b/per_scene/colorshape/qwen2.5-vl-7b-instruct__hard.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..00b45620c88acf2c73cd5f1a7f52c0114a883a85 --- /dev/null +++ b/per_scene/colorshape/qwen2.5-vl-7b-instruct__hard.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:32b54abc77491098c6bdd15fe28accadcadadaacb918fc4b326e8cda34683bf4 +size 330099 diff --git a/per_scene/colorshape/qwen3-vl-32b-instruct__easy.csv.gz b/per_scene/colorshape/qwen3-vl-32b-instruct__easy.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..3b085beb23f1ef50b815594328dff1a3ff13907b --- /dev/null +++ b/per_scene/colorshape/qwen3-vl-32b-instruct__easy.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:74fb20fbabc1e91c101f53b7250aa334e5a28be843133147af110fe5fa331c1a +size 310898 diff --git a/per_scene/colorshape/qwen3-vl-32b-instruct__hard.csv.gz b/per_scene/colorshape/qwen3-vl-32b-instruct__hard.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..b01b71ef866d31b7dc3c0914bb5629ef04f517a5 --- /dev/null +++ b/per_scene/colorshape/qwen3-vl-32b-instruct__hard.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a5855b7796f24bd13491968acecb80b17680ceed2ec54d336388dc878553ebef +size 335732 diff --git a/per_scene/colorshape/qwen3-vl-8b-instruct__easy.csv.gz b/per_scene/colorshape/qwen3-vl-8b-instruct__easy.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..a533e126e19792aefb80b83718114c65828e4362 --- /dev/null +++ b/per_scene/colorshape/qwen3-vl-8b-instruct__easy.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:70a63c81d4d462a191f49974bc54e39f06a4a6c43f92f2bd46d3089869aa7b6e +size 319153 diff --git a/per_scene/colorshape/qwen3-vl-8b-instruct__hard.csv.gz b/per_scene/colorshape/qwen3-vl-8b-instruct__hard.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..acb93ff707a2a5b4e59ae3b4c37d8b78197b7bad --- /dev/null +++ b/per_scene/colorshape/qwen3-vl-8b-instruct__hard.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:983913ed0defc96700613d8fdae2923d7706d1c38ff98ac53da5533c5c2f36bd +size 334924 diff --git a/per_scene/gpt-4.1__api_subset.csv.gz b/per_scene/gpt-4.1__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..8e3dee0f0dc7252e7347fc5997ccf113c023441c --- /dev/null +++ b/per_scene/gpt-4.1__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f8619b242719af8e2872a734111985e32c81e62367b2c40048500ad8cd7391f1 +size 328786 diff --git a/per_scene/gpt-4o__api_subset.csv.gz b/per_scene/gpt-4o__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..c3510004952fbeb42f90b8bbacc0c01dc7639ec7 --- /dev/null +++ b/per_scene/gpt-4o__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab32526e3f8bbce5202d97230c2b370876bf68427c746853d5262bd66f51b036 +size 341965 diff --git a/per_scene/gpt-5.5__api_subset.csv.gz b/per_scene/gpt-5.5__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..d3a3f513d3a8f279556318498e614926d08cfc22 --- /dev/null +++ b/per_scene/gpt-5.5__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5268ef2901cf992a14f41d97a25792d37e250e9fde255c3effb0bcdd0a171c4f +size 3571 diff --git a/per_scene/internvl3_5-38b-hf__api_subset.csv.gz b/per_scene/internvl3_5-38b-hf__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..bdc73b856da176275d86fe4c2ffe0d14733ba702 --- /dev/null +++ b/per_scene/internvl3_5-38b-hf__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31b45ba456e2e7033f7c303926c2e2f8d3fd38f92ea21408226ab0eafe61e103 +size 341290 diff --git a/per_scene/internvl3_5-38b-hf__full.csv.gz b/per_scene/internvl3_5-38b-hf__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..d2ba1fc849a2688c42a86101fcccedf874e4bd22 --- /dev/null +++ b/per_scene/internvl3_5-38b-hf__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:13ce723b49169ca58687c663f477b8b630c3278067263719456c0832017dcea9 +size 1201066 diff --git a/per_scene/internvl3_5-8b-hf__full.csv.gz b/per_scene/internvl3_5-8b-hf__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..4989b8051b7c6cb67e767de631ca03918c636619 --- /dev/null +++ b/per_scene/internvl3_5-8b-hf__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:53bf699d92faae1e74c31d5f2b3ebe13558196ba389e7ba594a2db0c95c93503 +size 1208168 diff --git a/per_scene/lfm2-vl-3b__full.csv.gz b/per_scene/lfm2-vl-3b__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..faccce424c3ead9c2dc5218fe6edbec4af53c94f --- /dev/null +++ b/per_scene/lfm2-vl-3b__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e31d81495b8c0e2ee9bc6a4a51c04c47bbe311740c5479fa5241821cbfe4cb82 +size 1193759 diff --git a/per_scene/llava-onevision-qwen2-7b-ov-hf__full.csv.gz b/per_scene/llava-onevision-qwen2-7b-ov-hf__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..a3df3390d794f69fd6bda67c478d592faa307b16 --- /dev/null +++ b/per_scene/llava-onevision-qwen2-7b-ov-hf__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd850c7085ec5077840932e763778e337268d66ee83b470e10615c1e4bd909a5 +size 1192160 diff --git a/per_scene/mistral-small-3.1-24b-instruct-2503__api_subset.csv.gz b/per_scene/mistral-small-3.1-24b-instruct-2503__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..e9d60042cefbf79fe3b02b0166f6f1bb8286571e --- /dev/null +++ b/per_scene/mistral-small-3.1-24b-instruct-2503__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:764495f7f1529f5bfc5e9c4c69eb2c425332cbf37b1eb0cea101a603e4c6ca52 +size 343144 diff --git a/per_scene/mistral-small-3.1-24b-instruct-2503__full.csv.gz b/per_scene/mistral-small-3.1-24b-instruct-2503__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..0c0ef128a42b00b5c78ee405cb9c5d68d31f2e96 --- /dev/null +++ b/per_scene/mistral-small-3.1-24b-instruct-2503__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:99fe4f2b37734da308592d29af3e19930c706c43796daf38fbde91233b8a46f1 +size 1205451 diff --git a/per_scene/models.csv b/per_scene/models.csv new file mode 100644 index 0000000000000000000000000000000000000000..b6b2c456e4a6ed5cec149982105fa1eb64c53f54 --- /dev/null +++ b/per_scene/models.csv @@ -0,0 +1,30 @@ +model,slug,scene_set,file,n_rows,n_base_scenes,response +GPT-4.1,gpt-4.1,api_subset,gpt-4.1__api_subset.csv.gz,6486,450,next-token letter probabilities +GPT-4o,gpt-4o,api_subset,gpt-4o__api_subset.csv.gz,6486,450,next-token letter probabilities +Qwen2.5-VL-72B,qwen2.5-vl-72b-instruct,full,qwen2.5-vl-72b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +Qwen2-VL-72B,qwen2-vl-72b-instruct,full,qwen2-vl-72b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +Qwen3-VL-32B,qwen3-vl-32b-instruct,full,qwen3-vl-32b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +Qwen2.5-VL-32B,qwen2.5-vl-32b-instruct,full,qwen2.5-vl-32b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +Qwen3-VL-8B,qwen3-vl-8b-instruct,full,qwen3-vl-8b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +InternVL3.5-8B,internvl3_5-8b-hf,full,internvl3_5-8b-hf__full.csv.gz,23040,1300,next-token letter probabilities +Qwen2.5-VL-7B,qwen2.5-vl-7b-instruct,full,qwen2.5-vl-7b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +Qwen2-VL-7B,qwen2-vl-7b-instruct,full,qwen2-vl-7b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +LLaVA-OV-7B,llava-onevision-qwen2-7b-ov-hf,full,llava-onevision-qwen2-7b-ov-hf__full.csv.gz,23040,1300,next-token letter probabilities +Qwen2.5-VL-3B,qwen2.5-vl-3b-instruct,full,qwen2.5-vl-3b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +Qwen2-VL-2B,qwen2-vl-2b-instruct,full,qwen2-vl-2b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +SmolVLM2-2.2B,smolvlm2-2.2b-instruct,full,smolvlm2-2.2b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +InternVL3.5-38B,internvl3_5-38b-hf,full,internvl3_5-38b-hf__full.csv.gz,23040,1300,next-token letter probabilities +Qwen3-VL-30B-A3B,qwen3-vl-30b-a3b-instruct,full,qwen3-vl-30b-a3b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +Mistral-Small-3.1-24B,mistral-small-3.1-24b-instruct-2503,full,mistral-small-3.1-24b-instruct-2503__full.csv.gz,23040,1300,next-token letter probabilities +Qwen3-VL-4B,qwen3-vl-4b-instruct,full,qwen3-vl-4b-instruct__full.csv.gz,23040,1300,next-token letter probabilities +LFM2-VL-3B,lfm2-vl-3b,full,lfm2-vl-3b__full.csv.gz,23040,1300,next-token letter probabilities +Phi-4-multimodal,phi-4-multimodal-instruct,full,phi-4-multimodal-instruct__full.csv.gz,23040,1300,next-token letter probabilities +Molmo2-8B,molmo2-8b,full,molmo2-8b__full.csv.gz,23040,1300,next-token letter probabilities +GPT-5.5,gpt-5.5,api_subset,gpt-5.5__api_subset.csv.gz,500,250,vote (parsed answer per rotation) +Qwen2-VL-7B,qwen2-vl-7b-instruct,api_subset,qwen2-vl-7b-instruct__api_subset.csv.gz,6486,450,next-token letter probabilities +Qwen2.5-VL-7B,qwen2.5-vl-7b-instruct,api_subset,qwen2.5-vl-7b-instruct__api_subset.csv.gz,6486,450,next-token letter probabilities +Qwen3-VL-8B,qwen3-vl-8b-instruct,api_subset,qwen3-vl-8b-instruct__api_subset.csv.gz,6486,450,next-token letter probabilities +Qwen3-VL-32B,qwen3-vl-32b-instruct,api_subset,qwen3-vl-32b-instruct__api_subset.csv.gz,6486,450,next-token letter probabilities +Mistral-Small-3.1-24B,mistral-small-3.1-24b-instruct-2503,api_subset,mistral-small-3.1-24b-instruct-2503__api_subset.csv.gz,6486,450,next-token letter probabilities +InternVL3.5-38B,internvl3_5-38b-hf,api_subset,internvl3_5-38b-hf__api_subset.csv.gz,6486,450,next-token letter probabilities +Qwen2.5-VL-72B,qwen2.5-vl-72b-instruct,api_subset,qwen2.5-vl-72b-instruct__api_subset.csv.gz,6486,450,next-token letter probabilities diff --git a/per_scene/molmo2-8b__full.csv.gz b/per_scene/molmo2-8b__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..9902b35101638de4e8898df8d428eb6b1a88e794 --- /dev/null +++ b/per_scene/molmo2-8b__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:296f72ec9ffa4bf6b46a6b36767b703479e87a83e579fe947fa171d250bdf43a +size 1202752 diff --git a/per_scene/phi-4-multimodal-instruct__full.csv.gz b/per_scene/phi-4-multimodal-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..faadf073b160865b279a19f3eac80968e5d9ab97 --- /dev/null +++ b/per_scene/phi-4-multimodal-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:470f903314f58ecd0bba36044d2acbaa058f249a25ca4c21f109d8e132064f1d +size 1199651 diff --git a/per_scene/photo_twins/gpt-4.1__keyframe.csv.gz b/per_scene/photo_twins/gpt-4.1__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..9f98cf02290e42facb7ed158c9d37e181f02d442 --- /dev/null +++ b/per_scene/photo_twins/gpt-4.1__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:61682324703f730f212b4d0a3b95747c974fe48526966cda29e3cfce294769b3 +size 121633 diff --git a/per_scene/photo_twins/internvl3_5-38b-hf__keyframe.csv.gz b/per_scene/photo_twins/internvl3_5-38b-hf__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..39ace43c2cb1e38fa89771cceebd0754767880ca --- /dev/null +++ b/per_scene/photo_twins/internvl3_5-38b-hf__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7016fbb92791cd076abf1b27eb9ff3c225afffbd4ab2beff08ad8411f58d1466 +size 331918 diff --git a/per_scene/photo_twins/internvl3_5-8b-hf__keyframe.csv.gz b/per_scene/photo_twins/internvl3_5-8b-hf__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..61e9f5afd8ce1b50784665fad63b17f681c61419 --- /dev/null +++ b/per_scene/photo_twins/internvl3_5-8b-hf__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:48f76b740d9c41258dd2b4b69ec80b8223bf81b85adbc34e3baaac7f5d57f76d +size 334170 diff --git a/per_scene/photo_twins/mistral-small-3.1-24b-instruct-2503__keyframe.csv.gz b/per_scene/photo_twins/mistral-small-3.1-24b-instruct-2503__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..f31e13e1ad58e34f13c34b0e95c5998f994ef809 --- /dev/null +++ b/per_scene/photo_twins/mistral-small-3.1-24b-instruct-2503__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa190aafb3aaf5beec5becf36c95ed388846b9923938a0c07bfc69831b5e191c +size 331212 diff --git a/per_scene/photo_twins/qwen2-vl-7b-instruct__keyframe.csv.gz b/per_scene/photo_twins/qwen2-vl-7b-instruct__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..1b94f9c90c80bfcdb8af778276e3e9724d495ca7 --- /dev/null +++ b/per_scene/photo_twins/qwen2-vl-7b-instruct__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b5d46a95b09ae6d61d2ffdd135a90a28defcb89353a4170b9a301c3d5ba1d243 +size 327911 diff --git a/per_scene/photo_twins/qwen2.5-vl-32b-instruct__keyframe.csv.gz b/per_scene/photo_twins/qwen2.5-vl-32b-instruct__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..5395ba0d676866098777129d29169757f653d3bb --- /dev/null +++ b/per_scene/photo_twins/qwen2.5-vl-32b-instruct__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5411fabd8f3a6ba1df2f9436b7cc910ee7215545e205923e64fd188e07a5d9e1 +size 330989 diff --git a/per_scene/photo_twins/qwen2.5-vl-72b-instruct__keyframe.csv.gz b/per_scene/photo_twins/qwen2.5-vl-72b-instruct__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..13d01c33b213e4ac9f0e30604684c01583c2ac4c --- /dev/null +++ b/per_scene/photo_twins/qwen2.5-vl-72b-instruct__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf1b01d3a857908618fe686b1e6d7919edaafb451b49f70c76fed58f6b34187c +size 331423 diff --git a/per_scene/photo_twins/qwen2.5-vl-7b-instruct__keyframe.csv.gz b/per_scene/photo_twins/qwen2.5-vl-7b-instruct__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..956aeb1969e30f62470e4d976c2f5f188f2db9d4 --- /dev/null +++ b/per_scene/photo_twins/qwen2.5-vl-7b-instruct__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64caaf28aa56281ea6884137b799c81d6aa1e7ee673e0ed75e9c64604aba9413 +size 330201 diff --git a/per_scene/photo_twins/qwen3-vl-32b-instruct__keyframe.csv.gz b/per_scene/photo_twins/qwen3-vl-32b-instruct__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..b7f12c7000022cb4cdbd14f4cc257e59000364e4 --- /dev/null +++ b/per_scene/photo_twins/qwen3-vl-32b-instruct__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3e53ae5e57dae368c9c0c66755e90d231d083e32e04fceeaed7a699ba6a6a75 +size 335175 diff --git a/per_scene/photo_twins/qwen3-vl-8b-instruct__blind.csv.gz b/per_scene/photo_twins/qwen3-vl-8b-instruct__blind.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..fe5ba3e06fdea2fb224e6ddbd503815e19e9b83b --- /dev/null +++ b/per_scene/photo_twins/qwen3-vl-8b-instruct__blind.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3f02d6001dd2141e060fa3f9dd767b9f35514c6fbe2d26ff28f2083746b2caa1 +size 76683 diff --git a/per_scene/photo_twins/qwen3-vl-8b-instruct__keyframe.csv.gz b/per_scene/photo_twins/qwen3-vl-8b-instruct__keyframe.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..6c0e591d7c36e523a1418e519757d34a0717df1e --- /dev/null +++ b/per_scene/photo_twins/qwen3-vl-8b-instruct__keyframe.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:833526fa72f19aa3b92229a7c1f8e3064f329637d63aa47a55c14555ba0f1791 +size 335268 diff --git a/per_scene/qwen2-vl-2b-instruct__full.csv.gz b/per_scene/qwen2-vl-2b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..a770f44b807ffc511d68adaf10c397d7deb69b9b --- /dev/null +++ b/per_scene/qwen2-vl-2b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7220c2f99999a4af45545a009f96ea4d5ad780191a839240ffd0c540cd4a6ede +size 1181417 diff --git a/per_scene/qwen2-vl-72b-instruct__full.csv.gz b/per_scene/qwen2-vl-72b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..9b73eb3d66512af626f678fbc831986383dedc34 --- /dev/null +++ b/per_scene/qwen2-vl-72b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:63dc6bc38d2757a98c5f0fba8274d61f7acdddd98bacb9dbf6e23d0e1b88a254 +size 1189277 diff --git a/per_scene/qwen2-vl-7b-instruct__api_subset.csv.gz b/per_scene/qwen2-vl-7b-instruct__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..17a8ab98eeac63b8d21b57713d8418fc155384ff --- /dev/null +++ b/per_scene/qwen2-vl-7b-instruct__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bfa3daf5db81a614fc1576488d0c0bc2f803e8977749763fb5c4be0c90e80e61 +size 337218 diff --git a/per_scene/qwen2-vl-7b-instruct__full.csv.gz b/per_scene/qwen2-vl-7b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..007afe4d9679c796e702b687f1ecd3fb722ffc79 --- /dev/null +++ b/per_scene/qwen2-vl-7b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2860760f5843b9cf842a71ae9d415accc26314fd05e2b44b99cac4c74b10d846 +size 1186488 diff --git a/per_scene/qwen2.5-vl-32b-instruct__full.csv.gz b/per_scene/qwen2.5-vl-32b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..63928a16838534535c62fa962588442e38a972e4 --- /dev/null +++ b/per_scene/qwen2.5-vl-32b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d04bc45a2b4ad9d7fd161849cc8b344574b57667760b89b9aa65c68a112c192 +size 1201006 diff --git a/per_scene/qwen2.5-vl-3b-instruct__full.csv.gz b/per_scene/qwen2.5-vl-3b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..ddf1f2cd2f441efe3e34da54b6a83c1d8d19c147 --- /dev/null +++ b/per_scene/qwen2.5-vl-3b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ba310baba3f94b844250cac16eef27755c5fb2c3a0a5cf29e1b0fc8cf93f73d +size 1184177 diff --git a/per_scene/qwen2.5-vl-72b-instruct__api_subset.csv.gz b/per_scene/qwen2.5-vl-72b-instruct__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..89884593c47c10457e3a31361c2b5698c8eddb93 --- /dev/null +++ b/per_scene/qwen2.5-vl-72b-instruct__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8cc6aaa2729565e4e6e24c7c18c440d4708b9ab4979208d8a8b7d63e6e930ea6 +size 339681 diff --git a/per_scene/qwen2.5-vl-72b-instruct__full.csv.gz b/per_scene/qwen2.5-vl-72b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..eb72c94f27524539d613745d2a926bc3e56330b5 --- /dev/null +++ b/per_scene/qwen2.5-vl-72b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:53da9c453205f0b2b1cc2a1fa6fd8588c7b1ac38c93f7f81ca6ceba0ff4a0e6c +size 1193935 diff --git a/per_scene/qwen2.5-vl-7b-instruct__api_subset.csv.gz b/per_scene/qwen2.5-vl-7b-instruct__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..a9735589ea983d6e180805c4c8aafee17d79d72d --- /dev/null +++ b/per_scene/qwen2.5-vl-7b-instruct__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:352b802a02d6f8f160073f512d09d0f495a891d8a269c215270973b1aeebaeea +size 339519 diff --git a/per_scene/qwen2.5-vl-7b-instruct__full.csv.gz b/per_scene/qwen2.5-vl-7b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..b20ad6c9e60a8f34307dc1d016bd1e0ee1663680 --- /dev/null +++ b/per_scene/qwen2.5-vl-7b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cda822f03fedfdb289807860a947cb7e22467f28a57b639e75d6db3760d3c9e +size 1193029 diff --git a/per_scene/qwen3-vl-30b-a3b-instruct__full.csv.gz b/per_scene/qwen3-vl-30b-a3b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..b9af7003db9896888879d6214e26ac1f787d9f45 --- /dev/null +++ b/per_scene/qwen3-vl-30b-a3b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:63b9cf1a523fe8d8a0e77d04e56eee80c3f1a093ca62f9b8c6e8c92c5b54276d +size 1198074 diff --git a/per_scene/qwen3-vl-32b-instruct__api_subset.csv.gz b/per_scene/qwen3-vl-32b-instruct__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..53e47b3e2a8bb5eefa063658834c4b2c09a83283 --- /dev/null +++ b/per_scene/qwen3-vl-32b-instruct__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dbae99800da48cc8ec9f3c55c51deae3d04184e82f8bcffdf67cb81ef116ee3a +size 348116 diff --git a/per_scene/qwen3-vl-32b-instruct__full.csv.gz b/per_scene/qwen3-vl-32b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..f8a0b8ec1542781e63fe507b819c92ac7408b7d3 --- /dev/null +++ b/per_scene/qwen3-vl-32b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e7b700aec6e709e40cedf520f0d20bf2f89097b7b75decb51ca87d3513c7a882 +size 1224629 diff --git a/per_scene/qwen3-vl-4b-instruct__full.csv.gz b/per_scene/qwen3-vl-4b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..cc533358ba7e212d5f73024ae7243cc6766f2659 --- /dev/null +++ b/per_scene/qwen3-vl-4b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:80f7ee727a9585cc3528959b5101974e8696e12ae764fb81c8fa61b7e1d39089 +size 1224239 diff --git a/per_scene/qwen3-vl-8b-instruct__api_subset.csv.gz b/per_scene/qwen3-vl-8b-instruct__api_subset.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..e8b055007a0390cc6d10f6e460da309220ea4d75 --- /dev/null +++ b/per_scene/qwen3-vl-8b-instruct__api_subset.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d1754bf39c47578d6de658fc3b636dec41d9b6837a77a8e1a2647113a2b3c08 +size 347478 diff --git a/per_scene/qwen3-vl-8b-instruct__full.csv.gz b/per_scene/qwen3-vl-8b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..7dde30dfc67d9c714542066a6d407b235909e305 --- /dev/null +++ b/per_scene/qwen3-vl-8b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f24707fcbdbf857b46ecb7067edda196a6b41f808a81a16c60b89f08634525ec +size 1222656 diff --git a/per_scene/smolvlm2-2.2b-instruct__full.csv.gz b/per_scene/smolvlm2-2.2b-instruct__full.csv.gz new file mode 100644 index 0000000000000000000000000000000000000000..4ffb6f5950e9da60d0d8733cfd1a9c99522622b2 --- /dev/null +++ b/per_scene/smolvlm2-2.2b-instruct__full.csv.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:076b46dd570067751475414c78af7187716f01a15d0fef18ae1ffc319a6a5c0f +size 1185469