Papers › Disentangled Representation Learning for Text-Video Retrieval

Disentangled Representation Learning for Text-Video Retrieval

14 Mar 2022arXiv:2203.07111archive 2025-07-28

Qiang Wang, Yanhao Zhang, Yun Zheng, Pan Pan, Xian-Sheng Hua

Cross-modality interaction is a critical component in Text-Video Retrieval (TVR), yet there has been little examination of how different influencing factors for computing interaction affect performance. This paper first studies the interaction paradigm in depth, where we find that its computation can be split into two terms, the interaction contents at different granularity and the matching function to distinguish pairs with the same semantics. We also observe that the single-vector representation and implicit intensive function substantially hinder the optimization. Based on these findings, we propose a disentangled framework to capture a sequential and hierarchical representation. Firstly, considering the natural sequential structure in both text and video inputs, a Weighted Token-wise Interaction (WTI) module is performed to decouple the content and adaptively exploit the pair-wise correlations. This interaction can form a better disentangled manifold for sequential inputs. Secondly, we introduce a Channel DeCorrelation Regularization (CDCR) to minimize the redundancy between the components of the compared vectors, which facilitate learning a hierarchical representation. We demonstrate the effectiveness of the disentangled representation on various benchmarks, e.g., surpassing CLIP4Clip largely by +2.9%, +3.1%, +7.9%, +2.3%, +2.8% and +6.5% R@1 on the MSR-VTT, MSVD, VATEX, LSMDC, AcitivityNet, and DiDeMo, respectively.

PaperPDFCodeCode Syntology ran

In Syntology 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="2203.07111")

Code

Syntology Ran 8 of 12 code samples harvested from 1 repository linked to this paper; 4 have no recorded run. Of those that ran: 2 ran · honoured contract; 1 ran · violated contract; 4 ran · our draft was wrong; 1 ran with no contract checked.

By repository: official repository: 12 samples from 1 repository, 8 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

foolwood/DRL officialmentioned on GitHubpytorchApache-2.0 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

12 samples harvested; 8 ran; 2 honoured the contract we drafted; 4 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · honoured contract
1ran · violated contract
4ran · our draft was wrong
1ran
4unverified

Licence: 0 of the 12 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 foolwood/DRL. “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.

swish foolwood/DRL/tvr/models/until_module.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 0f786c407fb1ee4c · report
all_gather foolwood/DRL/tvr/utils/comm.py official repository ran Apache-2.0 (permissive) · d741eb912b24880a · report
basic_clean foolwood/DRL/tvr/models/tokenization_clip.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 98f385d847636a3e · report
gelu foolwood/DRL/tvr/models/until_module.py official repository ran · honoured contract fingerprinted Apache-2.0 (permissive) · 50e1ffed03f484ec · report
get_pairs foolwood/DRL/tvr/models/tokenization_clip.py official repository ran · our draft was wrong Apache-2.0 (permissive) · d919ae32e5e4e616 · report
reduce_dict foolwood/DRL/tvr/utils/comm.py official repository ran · violated contract Apache-2.0 (permissive) · 2a3280757ce5a198 · report
warmup_linear foolwood/DRL/tvr/models/optimization.py official repository ran · honoured contract fingerprinted Apache-2.0 (permissive) · e0960afdb0d64aa7 · report
whitespace_clean foolwood/DRL/tvr/models/tokenization_clip.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 9542161e9640b858 · report
dataloader_msrvtt_test foolwood/DRL/tvr/dataloaders/data_dataloaders.py official repository unverified Apache-2.0 (permissive) · b6070984a76e9bf3 · report
dataloader_msrvtt_train foolwood/DRL/tvr/dataloaders/data_dataloaders.py official repository unverified Apache-2.0 (permissive) · 780f499e676f8267 · report
warmup_constant foolwood/DRL/tvr/models/optimization.py official repository unverified Apache-2.0 (permissive) · 59e4e730a9dc5dc6 · report
warmup_cosine foolwood/DRL/tvr/models/optimization.py official repository unverified Apache-2.0 (permissive) · 20b3d25ba1b796e5 · report

Tasks

Representation LearningRetrievalVideo Retrieval

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Retrieval DiDeMo DRL text-to-video Mean Rank 11.5 #25 of 40 Archive leaderboard report
Video Retrieval DiDeMo DRL text-to-video Median Rank 2.0 #25 of 40 Archive leaderboard report
Video Retrieval DiDeMo DRL text-to-video R@1 49.0 #25 of 40 Archive leaderboard report
Video Retrieval DiDeMo DRL text-to-video R@10 84.5 #25 of 40 Archive leaderboard report
Video Retrieval DiDeMo DRL text-to-video R@5 76.5 #25 of 40 Archive leaderboard report
Video Retrieval DiDeMo DRL video-to-text Mean Rank 7.9 #25 of 40 Archive leaderboard report
Video Retrieval DiDeMo DRL video-to-text Median Rank 2 #25 of 40 Archive leaderboard report
Video Retrieval DiDeMo DRL video-to-text R@1 49.9 #25 of 40 Archive leaderboard report
Video Retrieval DiDeMo DRL video-to-text R@10 83.3 #25 of 40 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL text-to-video Mean Rank 11.4 #11 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL text-to-video Median Rank 1 #11 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL text-to-video R@1 53.3 #11 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL text-to-video R@10 87.6 #11 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL text-to-video R@5 80.3 #11 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL video-to-text Mean Rank 7.6 #11 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL video-to-text Median Rank 1.0 #11 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL video-to-text R@1 56.2 #11 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL video-to-text R@10 87.4 #11 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA DRL video-to-text R@5 79.9 #11 of 63 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