Papers › Video-Text as Game Players: Hierarchical Banzhaf Interaction for Cross-Modal...

Video-Text as Game Players: Hierarchical Banzhaf Interaction for Cross-Modal Representation Learning

25 Mar 2023CVPR 2023 1arXiv:2303.14369archive 2025-07-28

Peng Jin, Jinfa Huang, Pengfei Xiong, Shangxuan Tian, Chang Liu, Xiangyang Ji, Li Yuan, Jie Chen

Contrastive learning-based video-language representation learning approaches, e.g., CLIP, have achieved outstanding performance, which pursue semantic interaction upon pre-defined video-text pairs. To clarify this coarse-grained global interaction and move a step further, we have to encounter challenging shell-breaking interactions for fine-grained cross-modal learning. In this paper, we creatively model video-text as game players with multivariate cooperative game theory to wisely handle the uncertainty during fine-grained semantic interaction with diverse granularity, flexible combination, and vague intensity. Concretely, we propose Hierarchical Banzhaf Interaction (HBI) to value possible correspondence between video frames and text words for sensitive and explainable cross-modal contrast. To efficiently realize the cooperative game of multiple video frames and multiple text words, the proposed method clusters the original video frames (text words) and computes the Banzhaf Interaction between the merged tokens. By stacking token merge modules, we achieve cooperative games at different semantic levels. Extensive experiments on commonly used text-video retrieval and video-question answering benchmarks with superior performances justify the efficacy of our HBI. More encouragingly, it can also serve as a visualization tool to promote the understanding of cross-modal interaction, which have a far-reaching impact on the community. Project page is available at https://jpthu17.github.io/HBI/.

PaperPDFConference PDFCodeCode 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 16 code samples harvested from 3 repositories linked to this paper: 12 ran (of which 7 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 1 where Syntology's instrument failed) · 4 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: 1 sample from 1 repository, 1 ran (1 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 15 samples from 2 repositories, 11 ran (10 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.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

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.

jpthu17/HBI officialmentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
jpthu17/dicosa mentioned on GitHubpytorchApache-2.0Syntology: 13 samples, 10 ran (1 instrument-failed) · 3 unverified report
jpthu17/diffusionret mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to DiffusionRet: Generative Text-Video Retrieval with Diffusion Model (arXiv:2303.09867; that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified) and Text-Video Retrieval with Disentangled Conceptualization and Set-to-Set Alignment (arXiv:2305.12218; that paper's own run record: 15 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 4 honoured, 0 violated, 3 with no contract checked; 8 where Syntology's instrument failed) · 8 unverified); nothing checks that they implement this paper's method. report
jpthu17/emcl mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

16 samples harvested: 12 ran (of which 7 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 1 where Syntology's instrument failed) · 4 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
11ran
4unverified

Licence: 0 of the 16 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”. For 4 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.

jpthu17/HBI official repository 1 sample, 1 ran (0 instrument-failed) · 0 unverified
BanzhafInteraction HBI/models/banzhaf.py ran Apache-2.0 (permissive) · e707b0fe8aa85751 · report fixture: signature retrieve_logits:Tensor/3/float, text_mask:Tensor/2/float, video_mask:Tensor/2/float, text_weight:Tensor/2/float, video_weight:Tensor/2/float · returned Tensor · level recorded 2026-09-24
jpthu17/dicosa community (archive-listed) 13 samples, 10 ran (1 instrument-failed) · 3 unverified
AllGather in 13 papers tvr/models/modeling.py ran Apache-2.0 (permissive) · 72bb773921596f80 · report fixture: returned AllGather · level recorded 2026-09-22
AttentionPool2d in 14 papers tvr/models/modeling.py ran Apache-2.0 (permissive) · 9dbc40efee885693 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
Bottleneck in 98 papers tvr/models/modeling.py ran · metamorphic tier: deterministic fingerprinted Apache-2.0 (permissive) · d9e854326a29af63 · report fixture: returned Bottleneck · level recorded 2026-09-22
CrossEn tvr/models/modeling.py ran · metamorphic tier: deterministic fingerprinted Apache-2.0 (permissive) · 0303fad775cdc882 · report fixture: returned CrossEn · level recorded 2026-09-22
ModifiedResNet in 15 papers tvr/models/modeling.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 09e8fef38647435a · report fixture: returned ModifiedResNet · level recorded 2026-09-22
ResidualAttentionBlock in 38 papers tvr/models/modeling.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 99083d92c6568c29 · report fixture: returned ResidualAttentionBlock · level recorded 2026-09-22
Transformer in 173 papers tvr/models/modeling.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 80385a1374622ebf · report fixture: returned Transformer · level recorded 2026-09-22
VisualTransformer in 12 papers tvr/models/modeling.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · bab52ac82134c899 · report fixture: returned VisualTransformer · level recorded 2026-09-22
_download tvr/models/modeling.py ran · our draft was wrong Apache-2.0 (permissive) · 95fdd52bfeffad7d · report fixture: signature url:str/0/str, root:str/0/str · returned str · level recorded 2026-09-22
available_models in 6 papers tvr/models/modeling.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 3a4824e9630d22f9 · report fixture: returned list · level recorded 2026-09-22
CLIP in 25 papers tvr/models/modeling.py unverified Apache-2.0 (permissive) · aacf9b08feb3ea59 · report
DiCoSA tvr/models/modeling.py unverified Apache-2.0 (permissive) · d285823632f22684 · report
convert_weights tvr/models/modeling.py unverified Apache-2.0 (permissive) · 56babc05b4c255dd · report
jpthu17/emcl community (archive-listed) 2 samples, 1 ran (0 instrument-failed) · 1 unverified
Emcl video_retrieval/EMCL-Net/tvr/models/modeling.py ran fingerprinted MIT (permissive) · 2eac2ed706ba48cd · report fixture: signature embds:Tensor/2/float, if_train:bool/0/bool · returned Tensor · level recorded 2026-09-24
EMCL video_retrieval/EMCL-Net/tvr/models/modeling.py unverified MIT (permissive) · 3fe3a6ee59bad490 · 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. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

3 assemblies on this paper: 3 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

3 sibling assemblies for Hierarchical Banzhaf Interaction (HBI) (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Hierarchical Banzhaf Interaction (HBI) (graph method: Hierarchical Banzhaf Interaction (HBI)). 15 bricks; individually on a fixture: 12 ran (of which 8 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. DiCoSA entry · level 0 · unverified · d285823632f22684
  2. CLIP helper · level 0 · unverified · aacf9b08feb3ea59
  3. CrossEn helper · level 1 · ran · 0303fad775cdc882
  4. LayerNorm helper · level 1 · ran · e296840ad4106ad9
  5. convert_weights helper · level 0 · unverified · 56babc05b4c255dd
  6. ModifiedResNet helper · level 1 · ran · 09e8fef38647435a
  7. VisualTransformer helper · level 1 · ran · bab52ac82134c899
  8. _download helper · level 1 · ran · our draft was wrong · 95fdd52bfeffad7d
  9. available_models helper · level 1 · ran · 3a4824e9630d22f9
  10. AttentionPool2d helper · level 1 · ran · 9dbc40efee885693
  11. Bottleneck helper · level 1 · ran · d9e854326a29af63
  12. Transformer model · level 1 · ran · 80385a1374622ebf
  13. ResidualAttentionBlock model · level 1 · ran · 99083d92c6568c29
  14. QuickGELU helper · level 1 · ran · 8c5c4de012c0aab9
  15. AllGather helper · level 1 · ran · 72bb773921596f80

harvested Assembly for Hierarchical Banzhaf Interaction (HBI) (graph method: Hierarchical Banzhaf Interaction (HBI)). 16 bricks; individually on a fixture: 13 ran (of which 8 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. EMCL entry · level 0 · unverified · 3fe3a6ee59bad490
  2. CLIP helper · level 0 · unverified · aacf9b08feb3ea59
  3. CrossEn helper · level 1 · ran · 0303fad775cdc882
  4. Emcl helper · level 1 · ran · 2eac2ed706ba48cd
  5. LayerNorm helper · level 1 · ran · e296840ad4106ad9
  6. convert_weights helper · level 0 · unverified · 56babc05b4c255dd
  7. ModifiedResNet helper · level 1 · ran · 09e8fef38647435a
  8. VisualTransformer helper · level 1 · ran · bab52ac82134c899
  9. _download helper · level 1 · ran · our draft was wrong · 95fdd52bfeffad7d
  10. available_models helper · level 1 · ran · 3a4824e9630d22f9
  11. AttentionPool2d helper · level 1 · ran · 9dbc40efee885693
  12. Bottleneck helper · level 1 · ran · d9e854326a29af63
  13. Transformer helper · level 1 · ran · 80385a1374622ebf
  14. ResidualAttentionBlock helper · level 1 · ran · 99083d92c6568c29
  15. QuickGELU helper · level 1 · ran · 8c5c4de012c0aab9
  16. AllGather helper · level 1 · ran · 72bb773921596f80

harvested Assembly for Hierarchical Banzhaf Interaction (HBI) (graph method: Hierarchical Banzhaf Interaction (HBI)). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. BanzhafInteraction entry · level 1 · ran · e707b0fe8aa85751

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="2303.14369") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Contrastive LearningQuestion AnsweringRepresentation LearningRetrievalVideo Question AnsweringVideo RetrievalVisual Question Answering (VQA)

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Question Answering MSRVTT-QA HBI Accuracy 46.2 #8 of 14 Archive leaderboard report
Video Retrieval ActivityNet HBI text-to-video Mean Rank 6.6 #23 of 31 Archive leaderboard report
Video Retrieval ActivityNet HBI text-to-video Median Rank 2.0 #23 of 31 Archive leaderboard report
Video Retrieval ActivityNet HBI text-to-video R@1 42.2 #23 of 31 Archive leaderboard report
Video Retrieval ActivityNet HBI text-to-video R@10 84.6 #23 of 31 Archive leaderboard report
Video Retrieval ActivityNet HBI text-to-video R@5 73.0 #23 of 31 Archive leaderboard report
Video Retrieval ActivityNet HBI video-to-text Mean Rank 6.5 #23 of 31 Archive leaderboard report
Video Retrieval ActivityNet HBI video-to-text Median Rank 2.0 #23 of 31 Archive leaderboard report
Video Retrieval ActivityNet HBI video-to-text R@1 42.4 #23 of 31 Archive leaderboard report
Video Retrieval ActivityNet HBI video-to-text R@10 86.0 #23 of 31 Archive leaderboard report
Video Retrieval ActivityNet HBI video-to-text R@5 73.0 #23 of 31 Archive leaderboard report
Video Retrieval DiDeMo HBI text-to-video Mean Rank 12.1 #30 of 40 Archive leaderboard report
Video Retrieval DiDeMo HBI text-to-video Median Rank 2.0 #30 of 40 Archive leaderboard report
Video Retrieval DiDeMo HBI text-to-video R@1 46.9 #30 of 40 Archive leaderboard report
Video Retrieval DiDeMo HBI text-to-video R@10 82.7 #30 of 40 Archive leaderboard report
Video Retrieval DiDeMo HBI text-to-video R@5 74.9 #30 of 40 Archive leaderboard report
Video Retrieval DiDeMo HBI video-to-text Mean Rank 8.7 #30 of 40 Archive leaderboard report
Video Retrieval DiDeMo HBI video-to-text Median Rank 2.0 #30 of 40 Archive leaderboard report
Video Retrieval DiDeMo HBI video-to-text R@1 46.2 #30 of 40 Archive leaderboard report
Video Retrieval DiDeMo HBI video-to-text R@10 82.7 #30 of 40 Archive leaderboard report
Video Retrieval DiDeMo HBI video-to-text R@5 73.0 #30 of 40 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI text-to-video Mean Rank 12.0 #24 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI text-to-video Median Rank 2.0 #24 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI text-to-video R@1 48.6 #24 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI text-to-video R@10 83.4 #24 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI text-to-video R@5 74.6 #24 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI video-to-text Mean Rank 8.9 #24 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI video-to-text Median Rank 2.0 #24 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI video-to-text R@1 46.8 #24 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI video-to-text R@10 84.3 #24 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA HBI video-to-text R@5 74.3 #24 of 63 Archive leaderboard report
Visual Question Answering (VQA) MSRVTT-QA HBI Accuracy 0.462 #11 of 34 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

CLIP

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