Papers › ST-Adapter: Parameter-Efficient Image-to-Video Transfer Learning

ST-Adapter: Parameter-Efficient Image-to-Video Transfer Learning

27 Jun 2022arXiv:2206.13559archive 2025-07-28

Junting Pan, Ziyi Lin, Xiatian Zhu, Jing Shao, Hongsheng Li

Capitalizing on large pre-trained models for various downstream tasks of interest have recently emerged with promising performance. Due to the ever-growing model size, the standard full fine-tuning based task adaptation strategy becomes prohibitively costly in terms of model training and storage. This has led to a new research direction in parameter-efficient transfer learning. However, existing attempts typically focus on downstream tasks from the same modality (e.g., image understanding) of the pre-trained model. This creates a limit because in some specific modalities, (e.g., video understanding) such a strong pre-trained model with sufficient knowledge is less or not available. In this work, we investigate such a novel cross-modality transfer learning setting, namely parameter-efficient image-to-video transfer learning. To solve this problem, we propose a new Spatio-Temporal Adapter (ST-Adapter) for parameter-efficient fine-tuning per video task. With a built-in spatio-temporal reasoning capability in a compact design, ST-Adapter enables a pre-trained image model without temporal knowledge to reason about dynamic video content at a small (~8%) per-task parameter cost, requiring approximately 20 times fewer updated parameters compared to previous work. Extensive experiments on video action recognition tasks show that our ST-Adapter can match or even outperform the strong full fine-tuning strategy and state-of-the-art video models, whilst enjoying the advantage of parameter efficiency. The code and model are available at https://github.com/linziyi96/st-adapter

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 10 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 10 samples from 1 repository, 10 ran (7 with no instrument failure; 3 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-24 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

linziyi96/st-adapter officialmentioned in papermentioned on GitHubpytorchMITSyntology: 10 samples, 10 ran (3 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

10 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · our draft was wrong
7ran

Licence: 3 of the 10 samples are pointer only, meaning Syntology does not serve that copy's text. For 3 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from linziyi96/st-adapter. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

linziyi96/st-adapter official repository 10 samples, 10 ran (3 instrument-failed) · 0 unverified
convert_to_one_hot in 10 papers video_dataset/mixup.py ran MIT (permissive) here; the stored copy is pointer only · 547be2babaca4859 · report fixture: signature targets:Tensor/1/int, num_classes:int/0/int, on_value:float/0/float, off_value:float/0/float · returned Tensor · level recorded 2026-09-25
crop_boxes in 5 papers video_dataset/transform.py ran fingerprinted MIT (permissive) here; the stored copy is pointer only · 157585dc6ae8a498 · report fixture: signature boxes:ndarray/2/float, x_offset:int/0/int, y_offset:int/0/int · returned ndarray · level recorded 2026-09-24
load_model in 479 papers utils.py ran MIT (permissive) · 319881416fb3b9dc · report fixture: signature args:Args/0/?, model_without_ddp:callable/0/?, optimizer:Adam/0/?, lr_sched:StepLR/0/?, loss_scaler:LossScaler/0/? · returned int · level recorded 2026-09-25
mixup_target in 22 papers video_dataset/mixup.py ran MIT (permissive) here; the stored copy is pointer only · fd3e5fd51b6d7296 · report fixture: signature target:Tensor/1/int, num_classes:int/0/int, lam:float/0/float, smoothing:float/0/float · returned Tensor · level recorded 2026-09-25
rand_bbox in 63 papers video_dataset/mixup.py ran MIT (permissive) · 830c7d87b9bb8311 · report fixture: signature img_shape:tuple/1/int, lam:float/0/float, margin:float/0/float, count:int/0/int · returned tuple · level recorded 2026-09-25
random_crop in 59 papers video_dataset/transform.py ran MIT (permissive) · 64ce021674e30869 · report fixture: signature images:Tensor/4/float, size:int/0/int, boxes:ndarray/2/float · returned tuple · level recorded 2026-09-25
random_short_side_scale_jitter in 6 papers video_dataset/transform.py ran MIT (permissive) · 472299265598d14e · report fixture: signature images:Tensor/4/float, min_size:int/0/int, max_size:int/0/int, boxes:ndarray/2/float, inverse_uniform_sampling:bool/0/bool, interpolation:str/0/str · returned tuple · level recorded 2026-09-25
shear_x in 43 papers video_dataset/rand_augment.py ran · our draft was wrong MIT (permissive) · e866e1196986f728 · report fixture: signature img:Image/0/?, factor:float/0/float, resample:int/0/int, fillcolor:tuple/1/int · returned Image · level recorded 2026-09-24
shear_y in 37 papers video_dataset/rand_augment.py ran · our draft was wrong MIT (permissive) · 7b2503c99c9447fb · report fixture: signature img:Image/0/?, factor:float/0/float, resample:int/0/int, fillcolor:tuple/1/int · returned Image · level recorded 2026-09-24
translate_x_rel in 27 papers video_dataset/rand_augment.py ran · our draft was wrong MIT (permissive) · 323759f906fbb394 · report fixture: signature img:Image/0/?, pct:float/0/float, resample:int/0/int, fillcolor:tuple/1/int · returned Image · level recorded 2026-09-24

Tasks

Action ClassificationAction RecognitionTemporal Action LocalizationTransfer LearningVideo Understandingparameter-efficient fine-tuning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Classification Kinetics-400 ST-Adapter (ViT-L, CLIP) Acc@1 87.2 #37 of 207 Archive leaderboard report
Action Classification Kinetics-400 ST-Adapter (ViT-L, CLIP) Acc@5 97.6 #37 of 207 Archive leaderboard report
Action Recognition Something-Something V2 ST-Adapter (ViT-L, CLIP) GFLOPs 8248 #24 of 123 Archive leaderboard report
Action Recognition Something-Something V2 ST-Adapter (ViT-L, CLIP) Top-1 Accuracy 72.3 #24 of 123 Archive leaderboard report
Action Recognition Something-Something V2 ST-Adapter (ViT-L, CLIP) Top-5 Accuracy 93.9 #24 of 123 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Adapter

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections