Papers › Video Graph Transformer for Video Question Answering

Video Graph Transformer for Video Question Answering

12 Jul 2022arXiv:2207.05342archive 2025-07-28

Junbin Xiao, Pan Zhou, Tat-Seng Chua, Shuicheng Yan

This paper proposes a Video Graph Transformer (VGT) model for Video Quetion Answering (VideoQA). VGT's uniqueness are two-fold: 1) it designs a dynamic graph transformer module which encodes video by explicitly capturing the visual objects, their relations, and dynamics for complex spatio-temporal reasoning; and 2) it exploits disentangled video and text Transformers for relevance comparison between the video and text to perform QA, instead of entangled cross-modal Transformer for answer classification. Vision-text communication is done by additional cross-modal interaction modules. With more reasonable video encoding and QA solution, we show that VGT can achieve much better performances on VideoQA tasks that challenge dynamic relation reasoning than prior arts in the pretraining-free scenario. Its performances even surpass those models that are pretrained with millions of external data. We further show that VGT can also benefit a lot from self-supervised cross-modal pretraining, yet with orders of magnitude smaller data. These results clearly demonstrate the effectiveness and superiority of VGT, and reveal its potential for more data-efficient pretraining. With comprehensive analyses and some heuristic observations, we hope that VGT can promote VQA research beyond coarse recognition/description towards fine-grained relation reasoning in realistic videos. Our code is available at https://github.com/sail-sg/VGT.

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

Syntology 14 code samples harvested from 1 repository linked to this paper: 9 ran (of which 5 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 14 samples from 1 repository, 9 ran (8 with no instrument failure; 1 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 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

sail-sg/vgt officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 14 samples, 9 ran (1 instrument-failed) · 5 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

14 samples harvested: 9 ran (of which 5 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · our draft was wrong
8ran
5unverified

Licence: 0 of the 14 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 sail-sg/vgt. “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”. For 3 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

sail-sg/vgt official repository 14 samples, 9 ran (1 instrument-failed) · 5 unverified
CMAtten model/vqa_model.py ran Apache-2.0 (permissive) · f3ff51937dd98c21 · report fixture: signature s1:Tensor/3/float, l1:Tensor/1/int, s2:Tensor/3/float, l2:Tensor/1/int · returned tuple · level recorded 2026-09-24
Embeddings in 15 papers model/vqa_model.py ran Apache-2.0 (permissive) · 5a6a727d5b0f9c6b · report fixture: returned Embeddings · level recorded 2026-09-22
EncoderVid model/vqa_model.py ran Apache-2.0 (permissive) · 66dcb62d12860979 · report fixture: returned EncoderVid · level recorded 2026-09-22
FFN in 37 papers model/vqa_model.py ran fingerprinted Apache-2.0 (permissive) · fad49f16c43d52df · report fixture: returned FFN · level recorded 2026-09-22
Graph in 18 papers model/vqa_model.py ran Apache-2.0 (permissive) · c7325ef7876ef60e · report fixture: returned Graph · level recorded 2026-09-22
GraphConvolution in 36 papers model/vqa_model.py ran Apache-2.0 (permissive) · 889a03d6a437ba1c · report fixture: returned GraphConvolution · level recorded 2026-09-22
MultiHeadSelfAttention in 17 papers model/vqa_model.py ran Apache-2.0 (permissive) · b091b2a22b3049e3 · report fixture: signature query:Tensor/3/float, key:Tensor/3/float, value:Tensor/3/float, mask:Tensor/2/float, head_mask:NoneType/0/?, output_attentions:bool/0/bool · returned tuple · level recorded 2026-09-24
TransformerBlock in 73 papers model/vqa_model.py ran fingerprinted Apache-2.0 (permissive) · b294b4902336fb78 · report fixture: signature x:Tensor/3/float, attn_mask:Tensor/2/float, head_mask:NoneType/0/?, output_attentions:bool/0/bool · returned tuple · level recorded 2026-09-24
get_mask in 119 papers
census: 9 implementations from 10 papers, 9 distinct outputs across 7 buckets, one shared input per bucket this sample: bucket 1 of 7, arg 1: rank 1, kind int, dtype int64
model/vqa_model.py ran · our draft was wrong Apache-2.0 (permissive) · 4b0473dade9942bc · report fixture: signature lengths:Tensor/1/int, max_length:int/0/int · returned Tensor · level recorded 2026-09-22
AModel model/vqa_model.py unverified Apache-2.0 (permissive) · 2f0df7b215389abb · report
Bert in 8 papers model/vqa_model.py unverified Apache-2.0 (permissive) · f64528547183d480 · report
Transformer in 173 papers model/vqa_model.py unverified Apache-2.0 (permissive) · 9d80cb2ca7b058a1 · report
VGT model/vqa_model.py unverified Apache-2.0 (permissive) · cdedc918d8a5dc8b · report
create_sinusoidal_embeddings in 5 papers model/vqa_model.py unverified Apache-2.0 (permissive) · 741ba8df50c491fc · 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 Video Graph Transformer (VGT) (graph method: Video Graph Transformer (VGT)). 14 bricks; individually on a fixture: 9 ran (of which 5 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified. The assembly has not been run end-to-end.

  1. VGT entry · level 0 · unverified · cdedc918d8a5dc8b
  2. AModel helper · level 0 · unverified · 2f0df7b215389abb
  3. CMAtten model · level 1 · ran · f3ff51937dd98c21
  4. Embeddings model · level 1 · ran · 5a6a727d5b0f9c6b
  5. EncoderVid model · level 1 · ran · 66dcb62d12860979
  6. Graph model · level 1 · ran · c7325ef7876ef60e
  7. Transformer model · level 0 · unverified · 9d80cb2ca7b058a1
  8. get_mask helper · level 1 · ran · our draft was wrong · 4b0473dade9942bc
  9. Bert helper · level 0 · unverified · f64528547183d480
  10. create_sinusoidal_embeddings helper · level 0 · unverified · 741ba8df50c491fc
  11. GraphConvolution model · level 1 · ran · 889a03d6a437ba1c
  12. TransformerBlock model · level 1 · ran · b294b4902336fb78
  13. FFN model · level 1 · ran · fad49f16c43d52df
  14. MultiHeadSelfAttention model · level 1 · ran · b091b2a22b3049e3

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="2207.05342") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Question AnsweringVideo Question AnsweringVisual Question Answering (VQA)

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Question Answering IntentQA VGT Accuarcy 51.3 #5 of 6 Archive leaderboard report
Video Question Answering IntentQA VGT CH 56.0 #5 of 6 Archive leaderboard report
Video Question Answering IntentQA VGT CW 51.4 #5 of 6 Archive leaderboard report
Video Question Answering IntentQA VGT TP&TN 47.6 #5 of 6 Archive leaderboard report
Video Question Answering NExT-QA VGT(PT) Accuracy 56.9 #42 of 47 Archive leaderboard report
Video Question Answering NExT-QA VGT Accuracy 55.0 #44 of 47 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

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGraph TransformerLabel SmoothingLapEigenLaplacian PELayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

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