Papers › RelViT: Concept-guided Vision Transformer for Visual Relational Reasoning

RelViT: Concept-guided Vision Transformer for Visual Relational Reasoning

24 Apr 2022ICLR 2022 4arXiv:2204.11167archive 2025-07-28

Xiaojian Ma, Weili Nie, Zhiding Yu, Huaizu Jiang, Chaowei Xiao, Yuke Zhu, Song-Chun Zhu, Anima Anandkumar

Reasoning about visual relationships is central to how humans interpret the visual world. This task remains challenging for current deep learning algorithms since it requires addressing three key technical problems jointly: 1) identifying object entities and their properties, 2) inferring semantic relations between pairs of entities, and 3) generalizing to novel object-relation combinations, i.e., systematic generalization. In this work, we use vision transformers (ViTs) as our base model for visual reasoning and make better use of concepts defined as object entities and their relations to improve the reasoning ability of ViTs. Specifically, we introduce a novel concept-feature dictionary to allow flexible image feature retrieval at training time with concept keys. This dictionary enables two new concept-guided auxiliary tasks: 1) a global task for promoting relational reasoning, and 2) a local task for facilitating semantic object-centric correspondence learning. To examine the systematic generalization of visual reasoning models, we introduce systematic splits for the standard HICO and GQA benchmarks. We show the resulting model, Concept-guided Vision Transformer (or RelViT for short) significantly outperforms prior approaches on HICO and GQA by 16% and 13% in the original split, and by 43% and 18% in the systematic split. Our ablation analyses also reveal our model's compatibility with multiple ViT variants and robustness to hyper-parameters.

PaperPDFConference PDFCodeCode 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="2204.11167")

Code

Syntology Ran 6 of 10 code samples harvested from 1 repository linked to this paper; 4 have no recorded run. Of those that ran: 2 ran · honoured contract; 3 ran · fixture could not drive it; 1 ran with no contract checked.

By repository: official repository: 10 samples from 1 repository, 6 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

NVlabs/RelViT officialmentioned in papermentioned on GitHubpytorch 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

10 samples harvested; 6 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
3ran · fixture could not drive it
1ran
4unverified

Licence: 10 of the 10 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 NVlabs/RelViT. “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.

Buffer NVlabs/RelViT/utils/relvit.py official repository ran · metamorphic tier: deterministic licence not identified · pointer only · 4df4c993c5727035 · report
comp_sample_prob NVlabs/RelViT/utils/relvit.py official repository ran · honoured contract fingerprinted licence not identified · pointer only · 0c638badfdf212a4 · report
compute_acc_gqa NVlabs/RelViT/train_gqa.py official repository ran · honoured contract fingerprinted licence not identified · pointer only · 332c5303b119ea46 · report
gather_score_label NVlabs/RelViT/train_gqa.py official repository ran · fixture could not drive it licence not identified · pointer only · b6e1619d9365b133 · report
gather_score_label NVlabs/RelViT/train_hico.py official repository ran · fixture could not drive it licence not identified · pointer only · 75e49409a0d5bc35 · report
token_level_esvit NVlabs/RelViT/utils/relvit.py official repository ran · fixture could not drive it licence not identified · pointer only · 0a8ca463ea529cb4 · report
MoCo NVlabs/RelViT/utils/relvit.py official repository unverified licence not identified · pointer only · 5b8d7f4ee9f98aa7 · report
RCL NVlabs/RelViT/utils/relvit.py official repository unverified licence not identified · pointer only · 3a43770048acb61a · report
dequeue_with_concept NVlabs/RelViT/utils/relvit.py official repository unverified licence not identified · pointer only · 979ae321e669e678 · report
enqueue_with_concept NVlabs/RelViT/utils/relvit.py official repository unverified licence not identified · pointer only · 8aed60786f3eb5e0 · report

Tasks

Human-Object Interaction DetectionObjectRetrievalSystematic GeneralizationVisual Question Answering (VQA)Visual ReasoningZero-Shot Human-Object Interaction Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Human-Object Interaction Detection HICO RelViT mAP 43.98 #4 of 8 Archive leaderboard report
Visual Question Answering (VQA) GQA RelViT Accuracy 65.54 #2 of 2 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

AttentionLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionTransformerVision Transformer

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