Papers › 12-in-1: Multi-Task Vision and Language Representation Learning

12-in-1: Multi-Task Vision and Language Representation Learning

5 Dec 2019CVPR 2020 6arXiv:1912.02315archive 2025-07-28

Jiasen Lu, Vedanuj Goswami, Marcus Rohrbach, Devi Parikh, Stefan Lee

Much of vision-and-language research focuses on a small but diverse set of independent tasks and supporting datasets often studied in isolation; however, the visually-grounded language understanding skills required for success at these tasks overlap significantly. In this work, we investigate these relationships between vision-and-language tasks by developing a large-scale, multi-task training regime. Our approach culminates in a single model on 12 datasets from four broad categories of task including visual question answering, caption-based image retrieval, grounding referring expressions, and multi-modal verification. Compared to independently trained single-task models, this represents a reduction from approximately 3 billion parameters to 270 million while simultaneously improving performance by 2.05 points on average across tasks. We use our multi-task framework to perform in-depth analysis of the effect of joint training diverse tasks. Further, we show that finetuning task-specific models from our single multi-task model can lead to further improvements, achieving performance at or above the state-of-the-art.

PaperPDFConference PDFCodeCode Syntology ran

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 20 code samples harvested from 2 repositories linked to this paper: 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. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 14 samples from 1 repository, 11 ran (9 with no instrument failure; 2 where Syntology's instrument failed); community (archive-listed): 6 samples from 1 repository, 4 ran (4 with no instrument failure; 0 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 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 15 samples that ran has no recorded date.

facebookresearch/vilbert-multi-task officialmentioned in papermentioned on GitHubpytorchMITSyntology: 14 samples, 11 ran (2 instrument-failed) · 3 unverified report
Cloud-CV/vilbert-multi-task mentioned on GitHubNOASSERTIONSyntology: not harvested report
jialinwu17/tmpimgs mentioned on GitHubpytorchSyntology: not harvested report
jiasenlu/vilbert_beta mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for... (arXiv:1908.02265; that paper's own run record: 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); nothing checks that they implement this paper's method. report
johntiger1/multitask_multimodal mentioned on GitHubpytorchMITSyntology: 6 samples, 4 ran (0 instrument-failed) · 2 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

20 samples harvested: 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. “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 · honoured contract
2ran · our draft was wrong
12ran
5unverified

Licence: 20 of the 20 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 2 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”. “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.

facebookresearch/vilbert-multi-task official repository 14 samples, 11 ran (2 instrument-failed) · 3 unverified
swish in 258 papers
census: 14 implementations from 37 papers, 2 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
vilbert/basebert.py ran · our draft was wrong fingerprinted MIT recorded; this copy not marked cleared · pointer only · 0f786c407fb1ee4c · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded, date not recorded
LoadLosses vilbert/task_utils.py ran MIT recorded; this copy not marked cleared · pointer only · 8a9074f29e11ae33 · report fixture: signature args:dict/0/?, task_cfg:dict/0/?, task_ids:list/1/str · returned dict · level recorded 2026-09-25
cached_path in 47 papers vilbert/utils.py ran MIT recorded; this copy not marked cleared · pointer only · 5731b925b4c10219 · report fixture: signature url_or_filename:str/0/str, cache_dir:str/0/str · returned str · level recorded 2026-09-25
deserialize_lmdb vilbert/datasets/concept_cap_dataset.py ran MIT recorded; this copy not marked cleared · pointer only · 7a3860e69193cd1e · report fixture: signature ds:tuple/1/bytes · returned dict · level recorded 2026-09-25
filename_to_url in 54 papers vilbert/utils.py ran MIT recorded; this copy not marked cleared · pointer only · a60fbe82ff2a8bc4 · report fixture: signature filename:str/0/str, cache_dir:str/0/str · returned tuple · level recorded 2026-09-25
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
vilbert/basebert.py ran · honoured contract fingerprinted MIT recorded; this copy not marked cleared · pointer only · fdc64f4c72036ae4 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
get_annotations in 13 papers vilbert/datasets/flickr_grounding_dataset.py ran MIT recorded; this copy not marked cleared · pointer only · 32d8e4939405d6e2 · report fixture: signature fn:str/0/str · returned dict · level recorded 2026-09-25
get_sentence_data vilbert/datasets/flickr_grounding_dataset.py ran MIT recorded; this copy not marked cleared · pointer only · 7310149bac03cbe2 · report fixture: signature fn:str/0/str · returned list · level recorded 2026-09-25
iou in 97 papers
census: 5 implementations from 7 papers, 5 distinct outputs across 5 buckets, one shared input per bucket
vilbert/datasets/flickr_grounding_dataset.py ran fingerprinted MIT recorded; this copy not marked cleared · pointer only · a1bd52099cee7c87 · report fixture: signature anchors:Tensor/2/float, gt_boxes:Tensor/2/float · returned Tensor · level recorded 2026-09-25
iou in 97 papers
census: 5 implementations from 7 papers, 5 distinct outputs across 5 buckets, one shared input per bucket
vilbert/datasets/concept_cap_dataset.py ran fingerprinted MIT recorded; this copy not marked cleared · pointer only · 9499ed1255d97132 · report fixture: signature anchors:ndarray/2/float, gt_boxes:ndarray/2/float · returned ndarray · level recorded 2026-09-25
url_to_filename in 73 papers vilbert/utils.py ran · our draft was wrong fingerprinted MIT recorded; this copy not marked cleared · pointer only · 5eff22fa0a651276 · report fixture: signature url:str/0/str, etag:str/0/str · returned str · level recorded 2026-09-22
ForwardModelsTrain vilbert/task_utils.py unverified MIT recorded; this copy not marked cleared · pointer only · 4c86b7a9d48a484c · report
ForwardModelsVal vilbert/task_utils.py unverified MIT recorded; this copy not marked cleared · pointer only · ce5a514305bf7e1b · report
load_tf_weights_in_bert in 135 papers vilbert/vilbert.py unverified MIT recorded; this copy not marked cleared · pointer only · baa5766f4566aafc · report
johntiger1/multitask_multimodal community (archive-listed) 6 samples, 4 ran (0 instrument-failed) · 2 unverified
compute_score_with_logits in 12 papers evaluation/eval_vqa.py ran fingerprinted MIT recorded; this copy not marked cleared · pointer only · 89bdd6decba2b415 · report fixture: signature logits:Tensor/2/float, labels:Tensor/2/float · returned Tensor · level recorded 2026-09-24
evaluate in 514 papers evaluation/eval_refer_expression.py ran MIT recorded; this copy not marked cleared · pointer only · 701dca7f22198e4f · report fixture: signature args:DummyArgs/0/?, model:callable/0/?, dataloader:list/1/tuple · returned tuple · level recorded 2026-09-24
instance_bce_with_logits in 7 papers evaluation/eval_refer_expression.py ran fingerprinted MIT recorded; this copy not marked cleared · pointer only · 2d56ebf472887b22 · report fixture: signature logits:Tensor/2/float, labels:Tensor/2/float · returned Tensor · level recorded 2026-09-24
lr_lambda_update evaluation/eval_vqa.py ran fingerprinted MIT recorded; this copy not marked cleared · pointer only · 50cecb34b46ee0b0 · report fixture: signature i_iter:int/0/int · returned float · level recorded 2026-09-24
evaluate in 514 papers evaluation/eval_coco_retrieval.py unverified MIT recorded; this copy not marked cleared · pointer only · 2454efa3ee5d7949 · report
evaluate in 514 papers evaluation/eval_concap_retrieval.py unverified MIT recorded; this copy not marked cleared · pointer only · cee4c7bd2fc134c9 · report

Tasks

10-shot image generationImage RetrievalQuestion AnsweringRepresentation LearningRetrievalVisual Question AnsweringVisual Question Answering (VQA)

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
10-shot image generation MEAD 12 12k 12 #1 of 1 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