Papers › Multitask Prompted Training Enables Zero-Shot Task Generalization

Multitask Prompted Training Enables Zero-Shot Task Generalization

15 Oct 2021ICLR 2022 4arXiv:2110.08207archive 2025-07-28

Victor Sanh, Albert Webson, Colin Raffel, Stephen H. Bach, Lintang Sutawika, Zaid Alyafeai, Antoine Chaffin, Arnaud Stiegler, Teven Le Scao, Arun Raja, Manan Dey, M Saiful Bari, Canwen Xu, Urmish Thakker, Shanya Sharma Sharma, Eliza Szczechla, Taewoon Kim, Gunjan Chhablani, Nihal Nayak, Debajyoti Datta, Jonathan Chang, Mike Tian-Jian Jiang, Han Wang, Matteo Manica, Sheng Shen, Zheng Xin Yong, Harshit Pandey, Rachel Bawden, Thomas Wang, Trishala Neeraj, Jos Rozen, Abheesht Sharma, Andrea Santilli, Thibault Fevry, Jason Alan Fries, Ryan Teehan, Tali Bers, Stella Biderman, Leo Gao, Thomas Wolf, Alexander M. Rush

Large language models have recently been shown to attain reasonable zero-shot generalization on a diverse set of tasks (Brown et al., 2020). It has been hypothesized that this is a consequence of implicit multitask learning in language models' pretraining (Radford et al., 2019). Can zero-shot generalization instead be directly induced by explicit multitask learning? To test this question at scale, we develop a system for easily mapping any natural language tasks into a human-readable prompted form. We convert a large set of supervised datasets, each with multiple prompts with diverse wording. These prompted datasets allow for benchmarking the ability of a model to perform completely held-out tasks. We fine-tune a pretrained encoder-decoder model (Raffel et al., 2020; Lester et al., 2021) on this multitask mixture covering a wide variety of tasks. The model attains strong zero-shot performance on several standard datasets, often outperforming models up to 16x its size. Further, our approach attains strong performance on a subset of tasks from the BIG-bench benchmark, outperforming models up to 6x its size. All trained models are available at https://github.com/bigscience-workshop/t-zero and all prompts are available at https://github.com/bigscience-workshop/promptsource.

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 15 code samples harvested from 4 repositories linked to this paper: 13 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 1 violated, 8 with no contract checked; 4 where Syntology's instrument failed) · 2 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 6 samples from 2 repositories, 6 ran (4 with no instrument failure; 2 where Syntology's instrument failed); community (archive-listed): 9 samples from 2 repositories, 7 ran (5 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.

bigscience-workshop/promptsource officialmentioned in papermentioned on GitHubSyntology: 3 samples, 3 ran (2 instrument-failed) · 0 unverified report
bigscience-workshop/t-zero officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
DevSinghSachan/art mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Questions Are All You Need to Train a Dense Passage Retriever (arXiv:2206.10658; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
facebookresearch/metaicl mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to UnifiedQA: Crossing Format Boundaries With a Single QA System (arXiv:2005.00700; that paper's own run record: 4 ran (of which 3 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 1 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
gonglinyuan/metro_t0 mentioned on GitHubpytorchSyntology: not harvested report
hojjat-mokhtarabadi/promptsource mentioned on GitHubApache-2.0Syntology: 3 samples, 2 ran (2 instrument-failed) · 1 unverified report
openbiolink/promptsource mentioned on GitHubApache-2.0Syntology: not harvested report
ukplab/arxiv2022-context-injection-stance mentioned on GitHubpytorchApache-2.0Syntology: 6 samples, 5 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

15 samples harvested: 13 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 1 violated, 8 with no contract checked; 4 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · violated contract
3ran · our draft was wrong
1ran; the contract check could not run
8ran
2unverified

Licence: 3 of the 15 samples are pointer only, meaning Syntology does not serve that copy's text. For 3 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. 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. “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.

bigscience-workshop/promptsource official repository 3 samples, 3 ran (2 instrument-failed) · 0 unverified
choice in 13 papers promptsource/templates.py ran · violated contract fingerprinted Apache-2.0 (permissive) · bb35fcd18dd842fb · report fixture: signature choices:list/1/int · returned int · level recorded 2026-09-22
highlight in 8 papers promptsource/templates.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 406e3e3a9886da66 · report fixture: signature input:str/0/str · returned str · level recorded 2026-09-22
most_frequent in 6 papers promptsource/templates.py ran; the contract check could not run (RAISES) · metamorphic tier: well formed fingerprinted Apache-2.0 (permissive) · 10c4b8cfda850b4c · report fixture: signature items:list/1/int · returned list · level recorded 2026-09-22
bigscience-workshop/t-zero official repository 3 samples, 3 ran (0 instrument-failed) · 0 unverified
feature_to_spec t0/seqio_tasks/utils.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · b7af70f17bfb5e07 · report fixture: signature feature:callable/0/?, length:bool/0/bool · returned TensorSpec · level recorded 2026-09-24
hf_dataset_to_tf_dataset t0/seqio_tasks/utils.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · de62b3eab371bdf6 · report fixture: signature dataset:MockHFDataset/0/? · returned _FlatMapDataset · level recorded 2026-09-24
strip_whitespace in 5 papers t0/seqio_tasks/tasks.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 11b64b0a85dfb964 · report fixture: signature output_or_target:str/0/str, example:NoneType/0/?, is_target:bool/0/bool · returned str · level recorded 2026-09-24
hojjat-mokhtarabadi/promptsource community (archive-listed) 3 samples, 2 ran (2 instrument-failed) · 1 unverified
removeHyphen promptsource/utils.py ran · our draft was wrong Apache-2.0 (permissive) · 0b1cb631e6e315b0 · report fixture: signature example:dict/0/? · returned dict · level recorded 2026-09-24
renameDatasetColumn promptsource/utils.py ran · our draft was wrong Apache-2.0 (permissive) · a77010010d56e618 · report fixture: signature dataset:Dataset/0/? · returned Dataset · level recorded 2026-09-24
get_dataset_builder promptsource/utils.py unverified Apache-2.0 (permissive) · e14d51a44fec9477 · report
ukplab/arxiv2022-context-injection-stance community (archive-listed) 6 samples, 5 ran (0 instrument-failed) · 1 unverified
fake_query src/retrieve/retrieve_t0pp_context.py ran Apache-2.0 (permissive) · 9b27d71357f65d09 · report fixture: signature prompts:list/1/str, model:callable/0/?, tokenizer:StandInTokenizer/0/? · returned list · level recorded 2026-09-24
get_attribution_stats src/analysis/analyse_attributions.py ran Apache-2.0 (permissive) · aa86817b5ef44488 · report fixture: signature frame:DataFrame/0/? · returned list · level recorded 2026-09-24
get_divergence src/analysis/analyse_topics.py ran Apache-2.0 (permissive) · 35a244de1ccdbabd · report fixture: signature D_p:str/0/str, D_q:str/0/str, vectorizer:CountVectorizer/0/? · returned float64 · level recorded 2026-09-24
get_indomain_splits src/analysis/count_dataset_splits.py ran Apache-2.0 (permissive) · dd07b1fdb8bc71a0 · report fixture: signature samples:DataFrame/0/?, seed:int/0/int, stratify:bool/0/bool, dev_test_size:float/0/float · returned tuple · level recorded 2026-09-24
jensen_shannon_divergence src/analysis/analyse_topics.py ran fingerprinted Apache-2.0 (permissive) · f296f02f0a3d87ac · report fixture: signature p:ndarray/1/float, q:ndarray/1/float · returned float64 · level recorded 2026-09-24
query in 11 papers src/retrieve/retrieve_t0pp_context.py unverified Apache-2.0 (permissive) · 279ad86d58c62ffb · report

Tasks

BenchmarkingDecoderLanguage ModellingPrompt EngineeringZero-shot Generalization

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Test

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