Papers › CAST: Cross-Attention in Space and Time for Video Action Recognition

CAST: Cross-Attention in Space and Time for Video Action Recognition

30 Nov 2023NeurIPS 2023 11arXiv:2311.18825archive 2025-07-28

DongHo Lee, Jongseo Lee, Jinwoo Choi

Recognizing human actions in videos requires spatial and temporal understanding. Most existing action recognition models lack a balanced spatio-temporal understanding of videos. In this work, we propose a novel two-stream architecture, called Cross-Attention in Space and Time (CAST), that achieves a balanced spatio-temporal understanding of videos using only RGB input. Our proposed bottleneck cross-attention mechanism enables the spatial and temporal expert models to exchange information and make synergistic predictions, leading to improved performance. We validate the proposed method with extensive experiments on public benchmarks with different characteristics: EPIC-KITCHENS-100, Something-Something-V2, and Kinetics-400. Our method consistently shows favorable performance across these datasets, while the performance of existing methods fluctuates depending on the dataset characteristics.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2311.18825")

Code

Syntology Ran 9 of 17 code samples harvested from 1 repository linked to this paper; 8 have no recorded run. Of those that ran: 1 ran · honoured contract; 8 ran with no contract checked.

By repository: named in the paper: 17 samples from 1 repository, 9 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

khu-vll/cast mentioned in papermentioned on GitHubpytorchNOASSERTION 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

17 samples harvested; 9 ran; 1 honoured the contract we drafted; 8 have no recorded run. Read from Syntology's graph 2026-09-25; that is when this build read the record, not when the samples ran.

1ran · honoured contract
8ran
8unverified

Licence: 17 of the 17 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from khu-vll/cast. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

Attention khu-vll/cast/models/bidir_modeling_crossattn.py named in the paper ran · metamorphic tier: invariant licence not identified · pointer only · b181227b8cc95f3d · report
CrossAttentionT2S khu-vll/cast/models/bidir_modeling_crossattn.py named in the paper ran licence not identified · pointer only · ef0019b3743b17db · report
LayerNorm khu-vll/cast/models/bidir_modeling_crossattn.py named in the paper ran · metamorphic tier: invariant fingerprinted licence not identified · pointer only · 353426ba46a4932c · report
composition_train_class_batch KHU-VLL/CAST/engine_for_compomodel.py named in the paper ran licence not identified · pointer only · 457807b46c3549fe · report
crop_clip KHU-VLL/CAST/util_tools/functional.py named in the paper ran no licence file found · pointer only · bc39b8ed5fefa7eb · report
cross_train_class_batch KHU-VLL/CAST/engine_for_onemodel.py named in the paper ran licence not identified · pointer only · 2604141544b5eec0 · report
get_resize_sizes KHU-VLL/CAST/util_tools/functional.py named in the paper ran no licence file found · pointer only · 2b53e457d5632b4e · report
get_sinusoid_encoding_table khu-vll/cast/models/bidir_modeling_crossattn.py named in the paper ran · honoured contract licence not identified · pointer only · 9b2250a3d13ed688 · report
resize_clip KHU-VLL/CAST/util_tools/functional.py named in the paper ran no licence file found · pointer only · 981936c73d3cdd12 · report
Adapter khu-vll/cast/models/bidir_modeling_crossattn.py named in the paper unverified licence not identified · pointer only · 857af2a7de44faaa · report
Block khu-vll/cast/models/bidir_modeling_crossattn.py named in the paper unverified licence not identified · pointer only · 1dd28bf10048de7d · report
CrossAttentionS2T khu-vll/cast/models/bidir_modeling_crossattn.py named in the paper unverified licence not identified · pointer only · 445bb3da9e1748a4 · report
STCrossTransformer khu-vll/cast/models/bidir_modeling_crossattn.py named in the paper unverified licence not identified · pointer only · d26fe19969fbc2c2 · report
build_dataset KHU-VLL/CAST/dataset/datasets.py named in the paper unverified licence not identified · pointer only · bc71b534fe126be4 · report
clip_vit_base_patch16_224 KHU-VLL/CAST/models/clip_modelling_finetune.py named in the paper unverified licence not identified · pointer only · 76274d93634ef7d9 · report
get_loss_scale_for_deepspeed KHU-VLL/CAST/engine_for_compomodel.py named in the paper unverified no licence file found · pointer only · 29416043c7035e7c · report
tensor_normalize KHU-VLL/CAST/dataset/kinetics.py named in the paper unverified no licence file found · pointer only · 14b3b13b2e7ee844 · report

Tasks

Action ClassificationAction RecognitionAction Recognition In VideosVideo Understanding

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Classification Kinetics-400 CAST(ViT-B/16) Acc@1 85.3 #56 of 207 Archive leaderboard report
Action Recognition EPIC-KITCHENS-100 CAST(ViT-B/16) Action@1 49.3 #10 of 32 Archive leaderboard report
Action Recognition EPIC-KITCHENS-100 CAST(ViT-B/16) Noun@1 60.9 #10 of 32 Archive leaderboard report
Action Recognition EPIC-KITCHENS-100 CAST(ViT-B/16) Verb@1 72.5 #10 of 32 Archive leaderboard report
Action Recognition Something-Something V2 CAST(ViT-B/16) Top-1 Accuracy 71.6 #27 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.

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