Papers › ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for...

ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks

6 Aug 2019NeurIPS 2019 12arXiv:1908.02265archive 2025-07-28

Jiasen Lu, Dhruv Batra, Devi Parikh, Stefan Lee

We present ViLBERT (short for Vision-and-Language BERT), a model for learning task-agnostic joint representations of image content and natural language. We extend the popular BERT architecture to a multi-modal two-stream model, pro-cessing both visual and textual inputs in separate streams that interact through co-attentional transformer layers. We pretrain our model through two proxy tasks on the large, automatically collected Conceptual Captions dataset and then transfer it to multiple established vision-and-language tasks -- visual question answering, visual commonsense reasoning, referring expressions, and caption-based image retrieval -- by making only minor additions to the base architecture. We observe significant improvements across tasks compared to existing task-specific models -- achieving state-of-the-art on all four tasks. Our work represents a shift away from learning groundings between vision and language only as part of task training and towards treating visual grounding as a pretrainable and transferable capability.

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 34 code samples harvested from 4 repositories linked to this paper: 10 ran (of which 6 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 2 where Syntology's instrument failed) · 24 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 34 samples from 4 repositories, 10 ran (8 with no instrument failure; 2 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 2 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.

Mehrab-Tanjim/enforce-reasoning mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
Mehrab-Tanjim/vilbert-rationalization mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
facebookresearch/vilbert-multi-task mentioned on GitHubpytorchMITSyntology: 1 sample, 0 ran · 1 unverified report
fuqianya/ViLBERT-Paddle mentioned on GitHubpaddleSyntology: not harvested report
hwanheelee1993/vilbertscore mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
jialinwu17/tmpimgs mentioned on GitHubpytorchSyntology: not harvested report
jiasenlu/vilbert_beta mentioned on GitHubpytorchSyntology: 31 samples, 10 ran (2 instrument-failed) · 21 unverified report
johntiger1/multitask_multimodal mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to 12-in-1: Multi-Task Vision and Language Representation Learning (arXiv:1912.02315; that paper's own run record: 15 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 1 honoured, 0 violated, 12 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
vmurahari3/visdial-bert mentioned on GitHubpytorchBSD-3-ClauseSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Large-scale Pretraining for Visual Dialog: A Simple State-of-the-Art Baseline (arXiv:1912.02379; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
zihaow123/unimm mentioned on GitHubpytorchGPL-3.0Syntology: not harvested report
allenai/allennlp-models pytorchApache-2.0Syntology: not harvested 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

34 samples harvested: 10 ran (of which 6 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 2 where Syntology's instrument failed) · 24 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.

2ran · our draft was wrong
8ran
24unverified

Licence: 34 of the 34 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 4 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. 1 sample whose link to this paper records no repository is listed under one of this paper's own repositories, because identical code was harvested from it for another 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.

Mehrab-Tanjim/enforce-reasoning community (archive-listed) 1 sample, 0 ran · 1 unverified
VILBertForVLTasks vilbert/vilbert.py unverified no licence file found · pointer only · 17f4b6d3fa815530 · report
Mehrab-Tanjim/vilbert-rationalization community (archive-listed) 1 sample, 0 ran · 1 unverified
BertForMultiModalPreTraining vilbert/vilbert.py unverified no licence file found · pointer only · 22866620f2187abf · report
facebookresearch/vilbert-multi-task community (archive-listed) 1 sample, 0 ran · 1 unverified
load_tf_weights_in_bert in 135 papers vilbert/vilbert.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own unverified licence of this copy not recorded · baa5766f4566aafc · report
jiasenlu/vilbert_beta community (archive-listed) 31 samples, 10 ran (2 instrument-failed) · 21 unverified
BertBiAttention vilbert/vilbert.py ran fingerprinted no licence file found · pointer only · 1de0db942902853c · report fixture: signature input_tensor1:Tensor/3/float, attention_mask1:Tensor/4/float, input_tensor2:Tensor/3/float, attention_mask2:Tensor/4/float, co_attention_mask:Tensor/4/float, use_co_attention_mask:bool/0/bool · returned tuple · level recorded 2026-09-24
BertConfig in 14 papers vilbert/vilbert.py ran no licence file found · pointer only · f1288247bde85379 · report fixture: returned BertConfig · level recorded 2026-09-22
BertImageIntermediate vilbert/vilbert.py ran fingerprinted no licence file found · pointer only · 2a8b07cb440f837e · report fixture: returned BertImageIntermediate · level recorded 2026-09-22
BertImagePooler vilbert/vilbert.py ran no licence file found · pointer only · 70328fa848560f96 · report fixture: returned BertImagePooler · level recorded 2026-09-22
BertImageSelfAttention vilbert/vilbert.py ran no licence file found · pointer only · 3014ed49281d7791 · report fixture: returned BertImageSelfAttention · level recorded 2026-09-22
BertSelfAttention in 29 papers vilbert/vilbert.py ran no licence file found · pointer only · d5fe4e5fe4ebd34d · report fixture: returned BertSelfAttention · level recorded 2026-09-22
BertTextPooler vilbert/vilbert.py ran fingerprinted no licence file found · pointer only · 83db7fd470dda623 · report fixture: signature hidden_states:Tensor/3/float · returned Tensor · level recorded 2026-09-24
SimpleClassifier vilbert/vilbert.py ran no licence file found · pointer only · c0c9f7ad53f1ee3c · report fixture: returned SimpleClassifier · level recorded 2026-09-22
cached_path in 47 papers vilbert/vilbert.py ran · our draft was wrong no licence file found · pointer only · 2f25a70a523a0c93 · report fixture: signature url_or_filename:str/0/str, cache_dir:str/0/str · returned str · level recorded 2026-09-22
split_s3_path vilbert/vilbert.py ran · our draft was wrong no licence file found · pointer only · 10f91d914125d917 · report fixture: signature url:str/0/str · returned tuple · level recorded 2026-09-22
BertAttention in 28 papers vilbert/vilbert.py unverified no licence file found · pointer only · 04726b79ddfafc13 · report
BertBiOutput vilbert/vilbert.py unverified no licence file found · pointer only · 14d944f974c60346 · report
BertConnectionLayer vilbert/vilbert.py unverified no licence file found · pointer only · d7baca9b9f2d749e · report
BertEncoder in 20 papers vilbert/vilbert.py unverified no licence file found · pointer only · e91ad90558db9297 · report
BertImageAttention vilbert/vilbert.py unverified no licence file found · pointer only · ee52132ca5f99c92 · report
BertImageEmbeddings vilbert/vilbert.py unverified no licence file found · pointer only · 3ff23adbea3f5089 · report
BertImageLayer vilbert/vilbert.py unverified no licence file found · pointer only · 724f4da180330352 · report
BertImageOutput vilbert/vilbert.py unverified no licence file found · pointer only · e001e03421a462ce · report
BertImagePredictionHead vilbert/vilbert.py unverified no licence file found · pointer only · 77ef570afdceb98e · report
BertImageSelfOutput vilbert/vilbert.py unverified no licence file found · pointer only · 692a09b0c54a0eac · report
BertImgPredictionHeadTransform vilbert/vilbert.py unverified no licence file found · pointer only · a4c47f3a59425f8f · report
BertLMPredictionHead in 10 papers vilbert/vilbert.py unverified no licence file found · pointer only · 95b9fa155e0b962d · report
BertLayer in 23 papers vilbert/vilbert.py unverified no licence file found · pointer only · c29c18ac8dff6a26 · report
BertModel in 16 papers vilbert/vilbert.py unverified no licence file found · pointer only · ee054708d97537c8 · report
BertPreTrainedModel in 7 papers vilbert/vilbert.py unverified no licence file found · pointer only · 4bd7a207408ee835 · report
BertPreTrainingHeads in 5 papers vilbert/vilbert.py unverified no licence file found · pointer only · 2003aba9b6ab5ca4 · report
BertPredictionHeadTransform in 10 papers vilbert/vilbert.py unverified no licence file found · pointer only · dc0c32124c0de4c8 · report
VILBertForVLTasks vilbert/vilbert.py unverified no licence file found · pointer only · 7267644b6d49c787 · report
get_from_cache vilbert/vilbert.py unverified no licence file found · pointer only · e326ca73d3b1b208 · report
s3_etag vilbert/vilbert.py unverified no licence file found · pointer only · 205dfcf49bf3f283 · report
s3_get vilbert/vilbert.py unverified no licence file found · pointer only · 2cca3cc33d3d7b1c · 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 ViLBERT (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for ViLBERT (graph method: ViLBERT). 41 bricks; individually on a fixture: 15 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 5 where Syntology's instrument failed) · 26 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. VILBertForVLTasks entry · level 0 · unverified · 7267644b6d49c787
  2. BertModel model · level 0 · unverified · ee054708d97537c8
  3. BertPreTrainedModel helper · level 0 · unverified · 4bd7a207408ee835
  4. BertPreTrainingHeads helper · level 0 · unverified · 2003aba9b6ab5ca4
  5. SimpleClassifier model · level 1 · ran · c0c9f7ad53f1ee3c
  6. BertEmbeddings helper · level 0 · unverified · 822402eb726689f9
  7. BertEncoder helper · level 0 · unverified · e91ad90558db9297
  8. BertImageEmbeddings helper · level 0 · unverified · 3ff23adbea3f5089
  9. BertImagePooler helper · level 1 · ran · 70328fa848560f96
  10. BertTextPooler helper · level 1 · ran · 83db7fd470dda623
  11. BertConfig helper · level 1 · ran · f1288247bde85379
  12. cached_path helper · level 1 · ran · our draft was wrong · 2f25a70a523a0c93
  13. load_tf_weights_in_bert helper · level 0 · unverified · baa5766f4566aafc
  14. BertImagePredictionHead helper · level 0 · unverified · 77ef570afdceb98e
  15. BertLMPredictionHead helper · level 0 · unverified · 95b9fa155e0b962d
  16. BertConnectionLayer model · level 0 · unverified · d7baca9b9f2d749e
  17. BertImageLayer model · level 0 · unverified · 724f4da180330352
  18. BertLayer model · level 0 · unverified · c29c18ac8dff6a26
  19. get_from_cache helper · level 0 · unverified · e326ca73d3b1b208
  20. BertImgPredictionHeadTransform helper · level 0 · unverified · a4c47f3a59425f8f
  21. BertPredictionHeadTransform helper · level 0 · unverified · dc0c32124c0de4c8
  22. BertBiAttention model · level 1 · ran · 1de0db942902853c
  23. BertBiOutput helper · level 0 · unverified · 14d944f974c60346
  24. BertImageIntermediate helper · level 1 · ran · 2a8b07cb440f837e
  25. BertImageOutput helper · level 0 · unverified · e001e03421a462ce
  26. BertIntermediate helper · level 1 · ran · c7123d4962da3767
  27. BertOutput helper · level 0 · unverified · fb6862928ef36d03
  28. BertImageAttention helper · level 0 · unverified · ee52132ca5f99c92
  29. BertAttention helper · level 0 · unverified · 04726b79ddfafc13
  30. http_get helper · level 0 · unverified · 18f49ada16168f38
  31. s3_etag helper · level 0 · unverified · 205dfcf49bf3f283
  32. s3_get helper · level 0 · unverified · 2cca3cc33d3d7b1c
  33. url_to_filename helper · level 1 · ran · our draft was wrong · 5eff22fa0a651276
  34. BertImageSelfAttention helper · level 1 · ran · 3014ed49281d7791
  35. BertImageSelfOutput helper · level 0 · unverified · 692a09b0c54a0eac
  36. BertSelfAttention helper · level 1 · ran · d5fe4e5fe4ebd34d
  37. BertSelfOutput helper · level 0 · unverified · 178a7c8f563f0beb
  38. s3_request helper · level 1 · ran · our draft was wrong · 7661139be003d514
  39. split_s3_path helper · level 1 · ran · our draft was wrong · 10f91d914125d917
  40. gelu helper · level 1 · ran · honoured contract · fdc64f4c72036ae4
  41. swish helper · level 2 · ran · our draft was wrong · 0f786c407fb1ee4c

harvested Assembly for ViLBERT (graph method: ViLBERT). 41 bricks; individually on a fixture: 15 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 5 where Syntology's instrument failed) · 26 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. VILBertForVLTasks entry · level 0 · unverified · 17f4b6d3fa815530
  2. BertModel model · level 0 · unverified · ee054708d97537c8
  3. BertPreTrainedModel helper · level 0 · unverified · 4bd7a207408ee835
  4. BertPreTrainingHeads helper · level 0 · unverified · 2003aba9b6ab5ca4
  5. SimpleClassifier helper · level 1 · ran · c0c9f7ad53f1ee3c
  6. BertEmbeddings helper · level 0 · unverified · 822402eb726689f9
  7. BertEncoder helper · level 0 · unverified · e91ad90558db9297
  8. BertImageEmbeddings helper · level 0 · unverified · 3ff23adbea3f5089
  9. BertImagePooler helper · level 1 · ran · 70328fa848560f96
  10. BertTextPooler helper · level 1 · ran · 83db7fd470dda623
  11. BertConfig helper · level 1 · ran · f1288247bde85379
  12. cached_path helper · level 1 · ran · our draft was wrong · 2f25a70a523a0c93
  13. load_tf_weights_in_bert helper · level 0 · unverified · baa5766f4566aafc
  14. BertImagePredictionHead helper · level 0 · unverified · 77ef570afdceb98e
  15. BertLMPredictionHead helper · level 0 · unverified · 95b9fa155e0b962d
  16. BertConnectionLayer helper · level 0 · unverified · d7baca9b9f2d749e
  17. BertImageLayer model · level 0 · unverified · 724f4da180330352
  18. BertLayer model · level 0 · unverified · c29c18ac8dff6a26
  19. get_from_cache helper · level 0 · unverified · e326ca73d3b1b208
  20. BertImgPredictionHeadTransform helper · level 0 · unverified · a4c47f3a59425f8f
  21. BertPredictionHeadTransform helper · level 0 · unverified · dc0c32124c0de4c8
  22. BertBiAttention model · level 1 · ran · 1de0db942902853c
  23. BertBiOutput helper · level 0 · unverified · 14d944f974c60346
  24. BertImageIntermediate helper · level 1 · ran · 2a8b07cb440f837e
  25. BertImageOutput helper · level 0 · unverified · e001e03421a462ce
  26. BertIntermediate helper · level 1 · ran · c7123d4962da3767
  27. BertOutput helper · level 0 · unverified · fb6862928ef36d03
  28. BertImageAttention helper · level 0 · unverified · ee52132ca5f99c92
  29. BertAttention helper · level 0 · unverified · 04726b79ddfafc13
  30. http_get helper · level 0 · unverified · 18f49ada16168f38
  31. s3_etag helper · level 0 · unverified · 205dfcf49bf3f283
  32. s3_get helper · level 0 · unverified · 2cca3cc33d3d7b1c
  33. url_to_filename helper · level 1 · ran · our draft was wrong · 5eff22fa0a651276
  34. BertImageSelfAttention helper · level 1 · ran · 3014ed49281d7791
  35. BertImageSelfOutput helper · level 0 · unverified · 692a09b0c54a0eac
  36. BertSelfAttention helper · level 1 · ran · d5fe4e5fe4ebd34d
  37. BertSelfOutput helper · level 0 · unverified · 178a7c8f563f0beb
  38. s3_request helper · level 1 · ran · our draft was wrong · 7661139be003d514
  39. split_s3_path helper · level 1 · ran · our draft was wrong · 10f91d914125d917
  40. gelu helper · level 1 · ran · honoured contract · fdc64f4c72036ae4
  41. swish helper · level 2 · ran · our draft was wrong · 0f786c407fb1ee4c

harvested Assembly for ViLBERT (graph method: ViLBERT). 40 bricks; individually on a fixture: 14 ran (of which 6 constructed an object rather than computing a result; 9 with no instrument failure: 1 honoured, 0 violated, 8 with no contract checked; 5 where Syntology's instrument failed) · 26 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. BertForMultiModalPreTraining entry · level 0 · unverified · 22866620f2187abf
  2. BertModel model · level 0 · unverified · ee054708d97537c8
  3. BertPreTrainedModel helper · level 0 · unverified · 4bd7a207408ee835
  4. BertPreTrainingHeads helper · level 0 · unverified · 2003aba9b6ab5ca4
  5. BertEmbeddings helper · level 0 · unverified · 822402eb726689f9
  6. BertEncoder helper · level 0 · unverified · e91ad90558db9297
  7. BertImageEmbeddings model · level 0 · unverified · 3ff23adbea3f5089
  8. BertImagePooler helper · level 1 · ran · 70328fa848560f96
  9. BertTextPooler helper · level 1 · ran · 83db7fd470dda623
  10. BertConfig helper · level 1 · ran · f1288247bde85379
  11. cached_path helper · level 1 · ran · our draft was wrong · 2f25a70a523a0c93
  12. load_tf_weights_in_bert helper · level 0 · unverified · baa5766f4566aafc
  13. BertImagePredictionHead model · level 0 · unverified · 77ef570afdceb98e
  14. BertLMPredictionHead model · level 0 · unverified · 95b9fa155e0b962d
  15. BertConnectionLayer model · level 0 · unverified · d7baca9b9f2d749e
  16. BertImageLayer model · level 0 · unverified · 724f4da180330352
  17. BertLayer model · level 0 · unverified · c29c18ac8dff6a26
  18. get_from_cache helper · level 0 · unverified · e326ca73d3b1b208
  19. BertImgPredictionHeadTransform helper · level 0 · unverified · a4c47f3a59425f8f
  20. BertPredictionHeadTransform helper · level 0 · unverified · dc0c32124c0de4c8
  21. BertBiAttention model · level 1 · ran · 1de0db942902853c
  22. BertBiOutput helper · level 0 · unverified · 14d944f974c60346
  23. BertImageIntermediate helper · level 1 · ran · 2a8b07cb440f837e
  24. BertImageOutput helper · level 0 · unverified · e001e03421a462ce
  25. BertIntermediate helper · level 1 · ran · c7123d4962da3767
  26. BertOutput helper · level 0 · unverified · fb6862928ef36d03
  27. BertImageAttention helper · level 0 · unverified · ee52132ca5f99c92
  28. BertAttention helper · level 0 · unverified · 04726b79ddfafc13
  29. http_get helper · level 0 · unverified · 18f49ada16168f38
  30. s3_etag helper · level 0 · unverified · 205dfcf49bf3f283
  31. s3_get helper · level 0 · unverified · 2cca3cc33d3d7b1c
  32. url_to_filename helper · level 1 · ran · our draft was wrong · 5eff22fa0a651276
  33. BertImageSelfAttention helper · level 1 · ran · 3014ed49281d7791
  34. BertImageSelfOutput helper · level 0 · unverified · 692a09b0c54a0eac
  35. BertSelfAttention helper · level 1 · ran · d5fe4e5fe4ebd34d
  36. BertSelfOutput helper · level 0 · unverified · 178a7c8f563f0beb
  37. s3_request helper · level 1 · ran · our draft was wrong · 7661139be003d514
  38. split_s3_path helper · level 1 · ran · our draft was wrong · 10f91d914125d917
  39. gelu helper · level 1 · ran · honoured contract · fdc64f4c72036ae4
  40. swish helper · level 2 · ran · our draft was wrong · 0f786c407fb1ee4c

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="1908.02265") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image RetrievalQuestion AnsweringReferring Expression ComprehensionRetrievalVisual Commonsense ReasoningVisual GroundingVisual Question AnsweringVisual Question Answering (VQA)Visual Reasoning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Visual Question Answering (VQA) A-OKVQA ViLBERT - VQA DA VQA Score 12.0 #9 of 15 Archive leaderboard report
Visual Question Answering (VQA) A-OKVQA ViLBERT - VQA MC Accuracy 42.1 #9 of 15 Archive leaderboard report
Visual Question Answering (VQA) A-OKVQA ViLBERT DA VQA Score 25.9 #11 of 15 Archive leaderboard report
Visual Question Answering (VQA) A-OKVQA ViLBERT MC Accuracy 41.5 #11 of 15 Archive leaderboard report
Visual Question Answering (VQA) A-OKVQA ViLBERT - OK-VQA DA VQA Score 9.2 #13 of 15 Archive leaderboard report
Visual Question Answering (VQA) A-OKVQA ViLBERT - OK-VQA MC Accuracy 34.1 #13 of 15 Archive leaderboard report
Visual Question Answering (VQA) VQA v2 test-dev ViLBERT Accuracy 70.55 #30 of 56 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

Introduced by this paper: ViLBERT

Absolute Position EncodingsAdamAttentionAttention DropoutBERTBPEDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUResidual ConnectionSoftmaxTransformerViLBERTWeight DecayWordPiece

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