SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.16727 · CVPR · 2023

VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking

Yu Qiao, Yali Wang, Zhan Tong, Limin Wang, Yi Wang, Zhiyu Zhao, Yinan He, Bingkun Huang

arXiv · PDF · Open in the Atlas

Code that ran

2 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 4 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2303.16727.json · badge: /api/ran/2303.16727.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
OpenGVLab/VideoMAEv2 marked official in Papers with Code
Papers with Code archive
2 ran · 4 no recorded run
FunctionStatusWhere it lives
CosAttention Ran
ran on a synthesized input · constructed an object; no result was computed
OpenGVLab/VideoMAEv2/models/modeling_pretrain.py
code served (permissive licence) · get_code("81776e1d67d5b65d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, CosAttention: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define ARGS dict with all parameters for CosAttention.__init__
ARGS = {
    'dim': 8,
    'num_heads': 2,
    'qkv_bias': False,
    'qk_scale': None,
    'attn_drop': 0.1,
    'proj_drop': 0.1,
    'attn_head_dim': None,
}

# Create an instance to verify it works
model = nn.Module()
cos_attn = nn.Module.__new__(type('CosAttention', (nn.Module,), {
    '__init__': lambda self, **kw: None
}))

# Actually instantiate properly by importing the class
# (In real usage, CosAttention would be imported)
# For this setup, we just validate ARGS has correct structure

# Verify ARGS maps all __init__ parameters
init_params = ['dim', 'num_heads', 'qkv_bias', 'qk_scale', 'attn_drop', 'proj_drop', 'attn_head_dim']
for param in init_params:
    assert param in ARGS, f"Missing parameter: {param}"
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type CosAttention — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a CosAttention output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
PatchEmbed Ran
ran on a synthesized input · constructed an object; no result was computed
OpenGVLab/VideoMAEv2/models/modeling_pretrain.py
code served (permissive licence) · get_code("8ccae2733dda5cc9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, PatchEmbed: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from timm.models.layers import to_2tuple

torch.manual_seed(0)

ARGS = {
    "img_size": 224,
    "patch_size": 16,
    "in_chans": 3,
    "embed_dim": 768,
    "num_frames": 16,
    "tubelet_size": 2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
timm, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PatchEmbed — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PatchEmbed output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
Block Not yet run
unverified (harvested, no recorded run)
OpenGVLab/VideoMAEv2/models/modeling_pretrain.py
code served (permissive licence) · get_code("ef3f731f47fa62cd")
How it ran

no invocation recorded.

PretrainVisionTransformer Not yet run
unverified (harvested, no recorded run)
OpenGVLab/VideoMAEv2/models/modeling_pretrain.py
code served (permissive licence) · get_code("77fa748016cdd5bb")
How it ran

no invocation recorded.

PretrainVisionTransformerDecoder Not yet run
unverified (harvested, no recorded run)
OpenGVLab/VideoMAEv2/models/modeling_pretrain.py
code served (permissive licence) · get_code("23510dbbfb59c729")
How it ran

no invocation recorded.

PretrainVisionTransformerEncoder Not yet run
unverified (harvested, no recorded run)
OpenGVLab/VideoMAEv2/models/modeling_pretrain.py
code served (permissive licence) · get_code("a7e14063cbbf7bfa")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 6 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Scale is the primary factor for building a powerful foundation model that could well generalize to a variety of downstream tasks. However, it is still challenging to train video foundation models with billions of parameters. This paper shows that video masked autoencoder (VideoMAE) is a scalable and general self-supervised pre-trainer for building video foundation models. We scale the VideoMAE in both model and data with a core design. Specifically, we present a dual masking strategy for efficient pre-training, with an encoder operating on a subset of video tokens and a decoder processing another subset of video tokens. Although VideoMAE is very efficient due to high masking ratio in encoder, masking decoder can still further reduce the overall computational cost. This enables the efficient pre-training of billion-level models in video. We also use a progressive training paradigm that involves an initial pre-training on a diverse multi-sourced unlabeled dataset, followed by a post-pre-training on a mixed labeled dataset. Finally, we successfully train a video ViT model with a billion parameters, which achieves a new state-of-the-art performance on the datasets of Kinetics (90.0% on K400 and 89.9% on K600) and Something-Something (68.7% on V1 and 77.0% on V2). In addition, we extensively verify the pre-trained video ViT models on a variety of downstream tasks, demonstrating its effectiveness as a general video representation learner.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2303.16727")
get_code_for_paper("2303.16727")
have("2303.16727")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2303.16727.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2303.16727.svg)](https://syntology.ai/paper/2303.16727)

Connect an agent — have() is free.