Papers › Ca2-VDM: Efficient Autoregressive Video Diffusion Model with Causal Generation and...

Ca2-VDM: Efficient Autoregressive Video Diffusion Model with Causal Generation and Cache Sharing

25 Nov 2024arXiv:2411.16375archive 2025-07-28

Kaifeng Gao, Jiaxin Shi, Hanwang Zhang, Chunping Wang, Jun Xiao, Long Chen

With the advance of diffusion models, today's video generation has achieved impressive quality. To extend the generation length and facilitate real-world applications, a majority of video diffusion models (VDMs) generate videos in an autoregressive manner, i.e., generating subsequent clips conditioned on the last frame(s) of the previous clip. However, existing autoregressive VDMs are highly inefficient and redundant: The model must re-compute all the conditional frames that are overlapped between adjacent clips. This issue is exacerbated when the conditional frames are extended autoregressively to provide the model with long-term context. In such cases, the computational demands increase significantly (i.e., with a quadratic complexity w.r.t. the autoregression step). In this paper, we propose Ca2-VDM, an efficient autoregressive VDM with Causal generation and Cache sharing. For causal generation, it introduces unidirectional feature computation, which ensures that the cache of conditional frames can be precomputed in previous autoregression steps and reused in every subsequent step, eliminating redundant computations. For cache sharing, it shares the cache across all denoising steps to avoid the huge cache storage cost. Extensive experiments demonstrated that our Ca2-VDM achieves state-of-the-art quantitative and qualitative video generation results and significantly improves the generation speed. Code is available: https://github.com/Dawn-LX/CausalCache-VDM

PaperPDFCodeCode Syntology ranMethod as code

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

The archive's official repository (dawn-lx/causalcache-vdm) was not harvested; the 21 samples below come from 3 repositories the archive does not mark official.

Syntology 21 code samples harvested from 3 repositories linked to this paper: 15 ran (of which 5 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 0 violated, 9 with no contract checked; 6 where Syntology's instrument failed) · 6 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 6 samples from 1 repository, 2 ran (1 with no instrument failure; 1 where Syntology's instrument failed); found in paper text by Syntology: 15 samples from 2 repositories, 13 ran (8 with no instrument failure; 5 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 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

dawn-lx/causalcache-vdm officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: not harvested report
dawn-lx/causal-videogen mentioned on GitHubpytorchApache-2.0Syntology: 6 samples, 2 ran (1 instrument-failed) · 4 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

21 samples harvested: 15 ran (of which 5 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 0 violated, 9 with no contract checked; 6 where Syntology's instrument failed) · 6 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.

4ran · our draft was wrong
2ran; the contract check could not run
9ran
6unverified

Licence: 5 of the 21 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 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

dawn-lx/causal-videogen community (archive-listed) 6 samples, 2 ran (1 instrument-failed) · 4 unverified
build_module in 5 papers opensora/registry.py ran · our draft was wrong Apache-2.0 (permissive) · 496f3cd2af0df2e1 · report fixture: signature module:dict/0/?, builder:SimpleRegistry/0/? · returned Linear · level recorded 2026-09-25
gather_forward_split_backward opensora/acceleration/communications.py ran Apache-2.0 (permissive) · 97c082db7efd0240 · report fixture: signature input_:Tensor/2/float, process_group:MockProcessGroup/0/?, dim:int/0/int, grad_scale:NoneType/0/? · returned Tensor · level recorded 2026-09-25
all_to_all in 7 papers opensora/acceleration/communications.py unverified Apache-2.0 (permissive) · b51f59051d16f7df · report
auto_grad_checkpoint opensora/acceleration/checkpoint.py unverified Apache-2.0 (permissive) · 20336f4981288332 · report
partial_causal_flash_attn opensora/models/causal_stdit2/attention.py unverified Apache-2.0 (permissive) · 5cbd445824659be5 · report
split_forward_gather_backward opensora/acceleration/communications.py unverified Apache-2.0 (permissive) · a771d32bab25e50b · report
songweige/TATS found in paper text by Syntology 10 samples, 9 ran (2 instrument-failed) · 1 unverified
Block in 252 papers
census: 5 implementations from 5 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
tats/tats_transformer.py ran MIT (permissive) · c36e0664ff375f5b · report fixture: signature x:Tensor/3/float, layer_past:NoneType/0/?, return_present:bool/0/bool · returned Tensor · level recorded 2026-09-24
CausalSelfAttention in 28 papers tats/tats_transformer.py ran fingerprinted MIT (permissive) · 0dcc3d2b5989bb48 · report fixture: signature x:Tensor/3/float, layer_past:NoneType/0/? · returned tuple · level recorded 2026-09-24
GPT in 24 papers tats/tats_transformer.py ran MIT (permissive) · 20e50a0db3aadd39 · report fixture: returned GPT · level recorded 2026-09-22
GPTConfig tats/tats_transformer.py ran MIT (permissive) · cd407b6b077bc83e · report fixture: returned GPTConfig · level recorded 2026-09-22
Identity in 49 papers
census: 10 implementations from 13 papers, 4 distinct outputs across 4 buckets, one shared input per bucket
tats/tats_transformer.py ran MIT (permissive) · f020bb1b02990123 · report fixture: returned Identity · level recorded 2026-09-22
Labelator tats/tats_transformer.py ran MIT (permissive) · b9f40f41b93875c6 · report fixture: returned Labelator · level recorded 2026-09-22
SOSProvider tats/tats_transformer.py ran MIT (permissive) · 6e1c0c9a43690fdb · report fixture: returned SOSProvider · level recorded 2026-09-22
accuracy in 872 papers
census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket
tats/tats_transformer.py ran; the contract check could not run (RAISES) MIT (permissive) · 085a09050f4f41dc · report fixture: signature output:Tensor/2/float, target:Tensor/1/int, topk:tuple/1/int · returned list · level recorded 2026-09-22
shift_dim tats/tats_transformer.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 791774067a83bf99 · report fixture: signature x:Tensor/4/float, src_dim:int/0/int, dest_dim:int/0/int, make_contiguous:bool/0/bool · returned Tensor · level recorded 2026-09-22
Net2NetTransformer tats/tats_transformer.py unverified MIT (permissive) · af840b748b65873d · report
universome/stylegan-v found in paper text by Syntology 5 samples, 4 ran (3 instrument-failed) · 1 unverified
cfg_to_args_str src/infra/utils.py ran · our draft was wrong no licence file found · pointer only · 2dc4f95b756cb2b3 · report fixture: signature cfg:DictConfig/0/?, use_dashes:bool/0/bool · returned str · level recorded 2026-09-25
construct_experiments_args src/infra/slurm_batch_launch.py ran no licence file found · pointer only · 2c35749b6eaa5bfe · report fixture: signature cfg:DictConfig/0/?, experiments_list:NoneType/0/?, suffix:str/0/str · returned list · level recorded 2026-09-25
num_gpus_to_mem src/infra/utils.py ran · our draft was wrong no licence file found · pointer only · f29fb27caa548814 · report fixture: signature num_gpus:int/0/int, mem_per_gpu:int/0/int · returned str · level recorded 2026-09-25
to_dict in 13 papers src/infra/slurm_batch_launch.py ran · our draft was wrong no licence file found · pointer only · 90b1bec775ebe652 · report fixture: signature cfg:dict/0/? · returned dict · level recorded 2026-09-25
compute_fvd src/metrics/frechet_video_distance.py unverified no licence file found · pointer only · 0bf504441d3b1c0e · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for Ca2-VDM (graph method: Ca2-VDM). 12 bricks; individually on a fixture: 11 ran (of which 6 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 1 violated, 8 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

  1. Net2NetTransformer entry · level 0 · unverified · af840b748b65873d
  2. GPT model · level 1 · ran · 20e50a0db3aadd39
  3. Identity helper · level 1 · ran · f020bb1b02990123
  4. Labelator helper · level 1 · ran · b9f40f41b93875c6
  5. SOSProvider helper · level 1 · ran · 6e1c0c9a43690fdb
  6. accuracy helper · level 1 · ran; the contract check could not run (RAISES) · 085a09050f4f41dc
  7. disabled_train helper · level 1 · ran · violated contract · 4cb732f513d69dfd
  8. shift_dim helper · level 1 · ran; the contract check could not run (RAISES) · 791774067a83bf99
  9. Block helper · level 1 · ran · c36e0664ff375f5b
  10. GPTConfig helper · level 1 · ran · cd407b6b077bc83e
  11. AbstractEncoder helper · level 1 · ran · b1104d8e9026bff1
  12. CausalSelfAttention helper · level 1 · ran · 0dcc3d2b5989bb48

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2411.16375") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

DenoisingVideo Generation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Diffusion

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